UBC Faculty Research and Publications

Serum IgG subclass levels and risk of exacerbations and hospitalizations in patients with COPD Leitao Filho, Fernando S; Ra, Seung W; Mattman, Andre; Schellenberg, Robert S; Criner, Gerard J; Woodruff, Prescott G; Lazarus, Stephen C; Albert, Richard; Connett, John E; Han, Meilan K; Martinez, Fernando J; Leung, Janice M; Paul Man, S. F; Aaron, Shawn D; Reed, Robert M; Sin, Don D Feb 14, 2018

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata


52383-12931_2018_Article_733.pdf [ 770.34kB ]
JSON: 52383-1.0363887.json
JSON-LD: 52383-1.0363887-ld.json
RDF/XML (Pretty): 52383-1.0363887-rdf.xml
RDF/JSON: 52383-1.0363887-rdf.json
Turtle: 52383-1.0363887-turtle.txt
N-Triples: 52383-1.0363887-rdf-ntriples.txt
Original Record: 52383-1.0363887-source.json
Full Text

Full Text

RESEARCH Open AccessSerum IgG subclass levels and risk ofexacerbations and hospitalizations inpatients with COPDFernando Sergio Leitao Filho1,2,3, Seung Won Ra1,4, Andre Mattman5, Robert S. Schellenberg1,2,3, Gerard J. Criner6,Prescott G. Woodruff7, Stephen C. Lazarus7, Richard Albert8, John E. Connett9, Meilan K. Han10,Fernando J. Martinez11, Janice M. Leung1,2, S. F. Paul Man1,2, Shawn D. Aaron12†, Robert M. Reed13†,Don D. Sin1,2*† and for the Canadian Respiratory Research Network (CRRN)AbstractBackground: The literature is scarce regarding the prevalence and clinical impact of IgG subclass deficiency inCOPD. We investigated the prevalence of IgG subclass deficiencies and their association with exacerbations andhospitalizations using subjects from two COPD cohorts.Methods: We measured IgG subclass levels using immunonephelometry in serum samples from participants enrolledin two previous COPD trials: Macrolide Azithromycin for Prevention of Exacerbations of COPD (MACRO; n = 976) andSimvastatin for the Prevention of Exacerbations in Moderate-to-Severe COPD (STATCOPE; n = 653). All samples werecollected from clinically stable participants upon entry into both studies. IgG subclass deficiency was diagnosed whenIgG subclass levels were below their respective lower limit of normal: IgG1 < 2.8 g/L; IgG2 < 1.15 g/L; IgG3 < 0.24 g/L;and IgG4 < 0.052 g/L. To investigate the impact of IgG subclass levels on time to first exacerbation or hospitalization,we log-transformed IgG levels and performed Cox regression models, with adjustments for confounders.Results: One or more IgG subclass deficiencies were found in 173 (17.7%) and 133 (20.4%) participants in MACRO andSTATCOPE, respectively. Lower IgG1 or IgG2 levels resulted in increased risk of exacerbations with adjusted hazard ratios (HR)of 1.30 (95% CI, 1.10–1.54, p< 0.01) and 1.19 (95% CI, 1.05–1.35, p< 0.01), respectively in the MACRO study, with STATCOPEyielding similar results. Reduced IgG1 or IgG2 levels were also associated with increased risk of hospitalizations: the adjustedHR for IgG1 and IgG2 was 1.52 (95% CI: 1.15–2.02, p< 0.01) and 1.33 (95% CI, 1.08–1.64, p< 0.01), respectively for the MACROstudy; in STATCOPE, only IgG2 was an independent predictor of hospitalization. In our multivariate Cox models, IgG3 andIgG4 levels did not result in significant associations for both outcomes in either MACRO or STATCOPE cohorts.Conclusions: Approximately 1 in 5 COPD patients had one or more IgG subclass deficiencies. Reduced IgG subclass levelswere independent risk factors for both COPD exacerbations (IgG1 and IgG2) and hospitalizations (IgG2) in two COPD cohorts.Trial registration: This study used serum samples from participants of the MACRO (NCT00325897) and STATCOPE(NCT01061671) trials.Keywords: IgG, IgG subclass deficiency, COPD, Exacerbation, Hospitalization* Correspondence: don.sin@hli.ubc.ca†Equal contributors1Centre for Heart Lung Innovation, St. Paul’s Hospital, Vancouver, BC V6Z 1Y6,Canada2Department of Medicine (Division of Respiratory Medicine), University ofBritish Columbia, Vancouver, BC, CanadaFull list of author information is available at the end of the article© The Author(s). 2018 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, andreproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link tothe Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver(http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.Leitao Filho et al. Respiratory Research  (2018) 19:30 https://doi.org/10.1186/s12931-018-0733-zBackgroundChronic obstructive pulmonary disease (COPD) is one ofthe leading causes of morbidity and mortality worldwide,projected to become the third leading cause of death in2020 [1]. The course of COPD is punctuated by episodes ofacute exacerbations, which result in worse quality of life,accelerated disease progression, and increased risk of mor-tality [1, 2]. Hospitalizations due to severe exacerbationsare associated with poor survival, with a 5-year mortalityrate of 55% based on a previous study [3]. Respiratory tractinfections are major triggers of COPD exacerbations [1, 4].There is increasing evidence that immunoglobulins (Ig)play a crucial role in preventing infections, particularlysinopulmonary infections which likely relate to the con-tinuous exposure of the anatomy involved to a vast rangeof pathogens [5–9]. Immunoglobulin G (IgG) is the pre-dominant Ig class, corresponding to 75–80% of total Iglevels in blood, and consisting of four subclasses (IgG1,IgG2, IgG3, and IgG4) [10, 11]. The primary mechanismsby which IgG inactivates or eliminates pathogens involveactivation of the complement system and opsonization ofpathogens [12]. These processes result in target cell lysisand enhanced phagocytosis, respectively [10, 13].Recently, we demonstrated that hypogammaglobuline-mia (defined as total IgG levels < 7.0 g/L) was present inapproximately 1 out of 4 patients with moderate to se-vere COPD [14]. More importantly, those patients withreduced IgG levels demonstrated 50% to 100% higherrisk for both exacerbations and hospitalizations. The re-lationship between various subclasses of IgG and risk ofexacerbations, however, remains unknown, as only a fewarticles with a number of limitations have described thefrequency of IgG subclass deficiencies in COPD[15–17]and no prior studies have evaluated their impact on hos-pitalizations. Considering that COPD is a heterogeneousdisease with several clinical phenotypes [18], we ana-lyzed samples from patients who were at increased riskfor exacerbations, and a second independent COPD co-hort was also used to replicate our findings.MethodsStudy designThis is a retrospective study which used samples from partic-ipants of two COPD cohorts: Macrolide Azithromycin forPrevention of Exacerbations of COPD (MACRO – First co-hort; ClinicalTrials.gov number: NCT00325897) [19] andSimvastatin for the Prevention of Exacerbations inModerate-to-Severe COPD (STATCOPE – Replication co-hort; ClinicalTrials.gov number: NCT01061671) [20]. Briefly,the MACRO trial tested azithromycin (250 mg daily), amacrolide antibiotic with immunomodulatory and antibac-terial effects, whereas the STATCOPE trial investigated sim-vastatin (40 mg daily), compared to placebo for theprevention of COPD exacerbations. Both trials were double-blinded and placebo-controlled intervention studies. All par-ticipants had a diagnosis of COPD, aged 40 years or older,were current or former smokers (≥ 10 pack-year smokinghistory), and had a forced expiratory volume in 1 s (FEV1) <80% of predicted values after bronchodilator use.Additionally, all subjects had to meet at least one of thefollowing inclusion criteria: use of supplemental oxygen, pre-vious treatment with systemic glucocorticoids or antibioticagents for respiratory problems, or a history of an emergencydepartment visit or hospitalization for COPD exacerbation.Exacerbation and hospitalization dataIn both cohorts, the number and severity of exacerbationsthat occurred during follow-up were captured. The follow-ing scale was used to grade the severity of exacerbations: 1)mild – treatment conducted at home, with or without con-tacting a health care provider; 2) moderate – treatment re-quiring a visit to the emergency department; 3) severe –treatment requiring hospitalization; and 4) very severe – inwhich patients needed intensive care unit (ICU) admission.COPD-related hospitalizations were identified by capturingall events graded as severe and very severe exacerbations.IgG measurementsBaseline blood samples were collected from clinicallystable participants upon entry into both studies. A total of978 and 654 samples from the MACRO and STATCOPEcohorts, respectively, were available. Two samples fromthe MACRO trial (azithromycin arm) and one samplefrom the STATCOPE trial (placebo arm) were of insuffi-cient quantity to allow measurements of all IgG subclassesand were thus not included in the analysis. In total, IgGlevels were measured in 1629 samples, 976 from partici-pants of MACRO (468 in the azithromycin arm; 508 inthe placebo arm) and 653 from participants of STAT-COPE (322 in the simvastatin arm; 331 in the placeboarm). Serum IgG subclass levels were measured usingBinding Site IgG subclass (IgG1–4) reagents (Birmingham,UK) by immunonephelometry on a Siemens BNII nephel-ometer (BNII; Erlangen, Germany). All samples wereprocessed in the clinical laboratory at St. Paul’s Hospital,Vancouver, British Columbia, Canada. The following nor-mal ranges for IgG subclasses in adults were used: IgG1,2.80–8.00 g/L; IgG2, 1.15–5.70 g/L; IgG3, 0.24–1.25 g/L;IgG4, 0.052–1.25 g/L [21]. These ranges were obtainedfrom a healthy population and include mean levels ±2standard deviations [22]. IgG subclass deficiency was diag-nosed when IgG levels were below their respective lowerlimit of normal for each subclass: IgG1 < 2.80 g/L, IgG2 <1.15 g/L, IgG3 < 0.24 g/L, and IgG4 < 0.052 g/L.Statistical analysisIgG subclass levels were not normally distributed, and thusare presented as median and interquartile range. CategoricalLeitao Filho et al. Respiratory Research  (2018) 19:30 Page 2 of 10variables are shown as number and percentages. We usedthe Mann-Whitney U test and the Pearson′s Chi-square testto compare the IgG subclass levels and the frequency of IgGsubclass deficiencies between the two cohorts, respectively.The Mann-Whitney U test was also applied to analyze theIgG subclass levels according to the exacerbation status(participants with zero or one exacerbation vs. participantswith two or more exacerbations) and the occurrence of hos-pitalizations during follow-up. Additionally, we comparedthe rates of COPD exacerbations in person-years (expressedas the number of acute exacerbations divided by the durationof follow-up in years) between participants with IgG subclasslevels below and above the lower limit of normal (LLN)using the Mann-Whitney U test.To investigate the effect of IgG levels on time to firstCOPD exacerbation or hospitalization, we applied Coxproportional-hazards regression models to obtain a hazardratio (HR) and its 95% confidence intervals (CI) associatedwith these outcomes for each IgG subclass. In our Coxmodels, we adjusted for covariates that were consideredbiologically relevant to the outcomes of interest (exacerba-tions and hospitalizations) or that presented with a p < 0.2at the univariate analysis. The following variables were in-cluded in our multivariate models: IgG subclass levels (log-transformed), study drug (azithromycin in the derivationcohort, or simvastatin in the validation cohort), age, gender,ethnicity, smoking status, FEV1 (% of predicted value),long-term oxygen therapy, inhaled corticosteroid therapy atbaseline, and history of systemic steroids in the year priorto enrollment. To investigate any possible effect of steroids(as described above) on IgG subclass levels, we also in-cluded an interaction term in our multivariate models. ForIgG subclasses that demonstrated significant associations inour time to event analyses, we also obtained the adjustedpredicted probabilities of COPD exacerbations or hospitali-zations (using logistic regression models) and calculated thePearson’s correlation coefficient between these probabilitiesand IgG-levels (log-transformed). SPSS version 22.0 forWindows (IBM Corp., Armonk, NY, USA) was used fordata analysis. The level of statistical significance was set atp < 0.05 for all tests.ResultsCharacteristics of participants and prevalence of IgG deficiencyThe details of participant characteristics at baseline arepresented in Table 1. In MACRO, the participants wereolder (65.5 ± 8.6 vs. 62.8 ± 8.4 years, p < 0.001), weremore likely to be Caucasians (82.6% vs. 77.9%, p = 0.02)and had slightly lower FEV1 (39.8 ± 15.7% vs. 41.5 ±17.8%, p = 0.049), and a higher rate of long-term domi-ciliary oxygen use (60.1% vs. 48.7%, p < 0.001) comparedwith participants enrolled in STATCOPE. Conversely,the prevalence of current smokers was higher in STAT-COPE (28.8% vs. 21.2%, p < 0.001).Total IgG and IgG subclass levels were similar betweenboth cohorts (Table 2). Overall, 306 (18.8%) participants inthe merged cohort (n= 1629) presented with at least one IgGsubclass deficiency (173 (17.7%) and 133 (20.4%) in MACROand STATCOPE, respectively)(Table 2). IgG3 deficiency wasthe most frequent, whereas IgG1 deficiency was the leastcommon. We also identified cases showing two or more IgGsubclasses deficiencies simultaneously. The majority ofparticipants identified with an IgG subclass deficiency werealso diagnosed with hypogammaglobulinemia (defined astotal IgG < 7.0 g/L), especially participants presenting withIgG1 (72/74, n= 97.3%) or IgG2 (69/93, 74.2%) deficiencies.No differences regarding the frequency of IgG subclass defi-ciencies between both MACRO and STATCOPE cohortswere observed (Additional file 1: Table S1). The median andinterquartile range for each IgG subclass according to thepresence or absence of the related IgG subclass deficiencyare shown in Additional file 2: Tables S2 (First and Replica-tion cohorts) and Additional file 3: Table S3 (merged cohort).IgG subclass levels according to exacerbation status andhospitalizationsIn both cohorts, we detected significantly lower IgG1 andIgG2 levels among participants who developed two or moreexacerbations during follow-up versus those who did nothave any exacerbations or developed only one exacerbation(Additional file 4: Figure S1). Similar results were also ob-served in both cohorts according to hospitalization status,with participants who required hospitalizations duringfollow-up presenting with lower IgG1 and IgG2 subclasslevels (Additional file 5: Figure S2).Given the low frequency of IgG subclass deficiency (<10%) for all IgG subclasses in both cohorts, we mergedboth cohorts (n = 1629) and compared the rates of COPDexacerbations per person-year between participantsaccording to the presence or not of IgG1–4 deficiencies.We observed significantly higher rates of exacerbations perperson-year among participants with IgG1 deficiency (2.9 ±5.46 vs. 1.48 ± 1.86, p < 0.001) and IgG2 deficiency (2.10 ±1.99 vs. 1.51 ± 2.18, p = 0.001) compared to their counter-parts with normal IgG1 and IgG2 levels (Fig. 1). Addition-ally, the frequency of IgG1 and IgG2 deficiency was alsosignificantly higher among recurrent exacerbators com-pared to participants who remained stable or presentedwith only one exacerbation during the follow-up: IgG1–43/644 (6.7%) vs. 31/985 (3.1%), p = 0.001; IgG2–47/644 (7.3%)vs. 46/985 (4.7%), p = 0.025.Impact of IgG subclass levels on time to first COPDexacerbation and hospitalizationThe adjusted HRs for having a COPD exacerbation in-creased significantly with progressively lower IgG1 andIgG2 levels in MACRO (HR: 1.30, 95% CI: 1.10–1.54, p =0.002; HR: 1.19, 95% CI: 1.05–1.35, p = 0.006, respectively)Leitao Filho et al. Respiratory Research  (2018) 19:30 Page 3 of 10Table 1 Baseline characteristics of participants in the MACRO and STATCOPE cohortsVariable MACRO (n = 976) STATCOPE (n = 653) P*Age, years 65.5 ± 8.6 62.8 ± 8.4 < 0.001Gender (Male: Female) – no. (%) 584 (59.8):392 (40.2) 381 (58.3):272(41.7) 0.55Ethnicity (Caucasian) – no. (%) 806 (82.6) 509 (77.9) 0.02BMI, kg/m2 27.7 ± 6.2 27.1 ± 6.8 0.062Current Smokers – no. (%) 207 (21.2) 188 (28.8) < 0.001FVC, % predicted 70.2 ± 18.1 71.1 ± 19.1 0.29FEV1, % predicted 39.8 ± 15.7 41.5 ± 17.8 0.049Long-term oxygen therapy – no. (%) 587 (60.1) 318 (48.7) < 0.001Systemic steroid use in previous year – no. (%) 822 (84.2) 550 (84.2) 0.99Inhaled steroid use – no. (%) 764 (78.3) 484 (74.1) 0.06IgG1 (g/L) 5.09 (2.45) 5.26 (2.70) 0.18IgG2 (g/L) 2.57 (1.72) 2.61 (1.64) 0.82IgG3 (g/L) 0.62 (0.49) 0.63 (0.51) 0.96IgG4 (g/L) 0.21 (0.28) 0.22 (0.32) 0.99Total IgG (g/L) 8.68 (3.72) 8.84 (4.00) 0.70Values expressed as means ± SD. IgG levels expressed as median (interquartile range). *Student’s t-test, Mann–Whitney U-test (for comparisons of IgG levels), orChi-square test, as appropriate. Definition of abbreviations: BMI body mass index, FVC forced vital capacity, FEV1 forced expiratory volume in one second, IgGimmunoglobulin GTable 2 Frequency of IgG subclass deficiency according to the presence or absence of associated hypogammaglobulinemia in themerged cohort (MACRO and STATCOPE cohorts combined)IgG abnormality Merged Cohort (n = 1629)IgG1 deficiency – no. (%) 74 (4.5%)With hypogammaglobulinemia – no. (%) 72 (4.4%)No hypogammaglobulinemia – no. (%) 2 (0.1%)IgG2 deficiency – no. (%) 93 (5.7%)With hypogammaglobulinemia – no. (%) 69 (4.2%)No hypogammaglobulinemia – no. (%) 24 (1.5%)IgG3 deficiency – no. (%) 124 (7.6%)With hypogammaglobulinemia – no. (%) 72 (4.4%)No hypogammaglobulinemia – no. (%) 52 (3.2%)IgG4 deficiency – no. (%) 114 (7.0%)With hypogammaglobulinemia – no. (%) 73 (4.5%)No hypogammaglobulinemia – no. (%) 41 (2.5%)One or more IgG subclass deficiency – no. (%) 306 (18.8%)With hypogammaglobulinemia – no. (%) 197 (12.1%)No hypogammaglobulinemia – no. (%) 109 (6.7%)Two IgG subclass deficiencies combined – no. (%) 47 (2.9%)With hypogammaglobulinemia – no. (%) 37 (2.3%)No hypogammaglobulinemia – no. (%) 10 (0.6%)Three IgG subclass deficiencies combined – no. (%)a 20 (1.2%)All IgG subclass deficiencies combined – no. (%)a 4 (0.2%)The normal range for IgG levels in adults used in this analysis were: IgG1, 2.8–8.0 g/L; IgG2, 1.15–5.70 g/L; IgG3, 0.24–1.25 g/L; IgG4, 0.052–1.250 g/L; total IgG,7.0–16.0 g/L. aAll participants were diagnosed with hypogammaglobulinemia (total IgG < 7.0 g/L).Leitao Filho et al. Respiratory Research  (2018) 19:30 Page 4 of 10(Table 3). Since the IgG subclass levels in our models wereexpressed using a negative base-2 logarithmic scale, forevery one-unit increase on this log scale, the IgG subclasslevels decreased by 50% (negative log22 = − 1). Thus, inMACRO, a 50% reduction of IgG1 levels was associatedwith a 30% higher adjusted HR for COPD exacerbations.A similar reduction in IgG2 levels resulted in a 19% higheradjusted HR for this outcome. These findings were repli-cated in STATCOPE, as the adjusted HRs related to COPDexacerbations were 1.25 (95% CI: 1.02–1.54, p = 0.035) and1.17 (95% CI: 1.01–1.36, p = 0.046) for IgG1 and IgG2, re-spectively. IgG1 and IgG2 levels were also associated withhospitalizations in MACRO: the adjusted HR for every 50%reduction of IgG1 level was 1.52 (95% CI: 1.15–2.02, p =0.004) and for IgG2 was 1.33 (95% CI: 1.08–1.64, p = 0.007)(Table 4). In STATCOPE, only decreased IgG2 levels re-sulted in a higher risk of hospitalizations (adjusted HR =1.43, 95% CI: 1.12–1.83, p = 0.004). We also detected an in-verse relationship between IgG levels (log-transformed) andthe adjusted predicted risk of COPD exacerbations (Fig. 2)and hospitalizations (Fig. 3) in MACRO and STATCOPEcohorts (p < 0.05 for both outcomes).In our multivariate Cox models, IgG3 and IgG4 levels didnot result in significant associations for both outcomes ineither MACRO or STATCOPE cohorts. For these two IgGsubclasses, we also calculated the HRs using both cohortscombined (n = 1629) and obtained similar HRs comparedFig. 1 Comparison of rates of COPD exacerbations in person-years between participants with IgG subclass deficiency and normal IgG subclasslevels in the merged dataset (MACRO and STATCOPE cohorts combined, n = 1629). Error bars represent 95% confidence intervalTable 3 Unadjusted and adjusted hazard ratios for COPDexacerbations according to IgG subclass levels in the MACROand STATCOPE cohortsMACRO – Time to first ExacerbationIgGa Univariate analysis Multivariate analysisbHazard Ratio (95% CI) P Hazard Ratio (95% CI) PIgG1 1.43 (1.23–1.67) < 0.001 1.30 (1.10–1.54) 0.002IgG2 1.29 (1.15–1.45) < 0.001 1.19 (1.05–1.35) 0.006IgG3 0.99 (0.88–1.11) 0.84 0.95 (0.85–1.07) 0.39IgG4 1.09 (1.02–1.17) 0.009 1.05 (0.98–1.13) 0.17STATCOPE – Time to first ExacerbationIgGa Hazard Ratio (95% CI) P Hazard Ratio (95% CI) PIgG1 1.45 (1.20–1.75) < 0.001 1.25 (1.02–1.54) 0.035IgG2 1.32 (1.15–1.52) < 0.001 1.17 (1.01–1.36) 0.046IgG3 1.10 (0.98–1.25) 0.11 0.99 (0.87–1.13) 0.90IgG4 1.01 (0.94–1.08) 0.76 0.96 (0.89–1.04) 0.33aIgG subclass levels expressed using a negative-log transformation (base 2), withone-unit increase on this log scale being equivalent to a 50% decrease of IgGsubclass levels. bAdjusted hazard ratios for IgG levels were calculated using a CoxProportional Hazards model with adjustments for the following covariates: studydrug (azithromycin vs. placebo in MACRO - First cohort; simvastatin vs. placebo inSTATCOPE - Replication cohort), age, gender, ethnicity (Caucasian vs.Non-Caucasian), FEV1 (% of predicted values), smoking status (current vs. formersmokers), oxygen dependence, inhaled corticosteroid use, and treatment withsystemic steroids in previous year. Legend: 95% CI = 95% Confidence IntervalLeitao Filho et al. Respiratory Research  (2018) 19:30 Page 5 of 10to analyzing the cohorts separately: IgG3 (exacerbations:HR = 0.97, 95% CI: 0.89–1.05, p = 0.53; hospitalizations:HR = 0.97, 95% CI: 0.84–1.12, p = 0.70); IgG4 (exacerba-tions: HR = 1.01, 95% CI: 0.96–1.06, p = 0.71; hospitaliza-tions: 1.02, 95% CI: 0.94–1.12, p = 0.59).We observed a trend for lower IgG1 subclass levelsamong participants who were treated with systemic ste-roids in the year prior to enrollment (MACRO - out-come: time to first exacerbation, p = 0.057; STATCOPE -outcome: time to first hospitalization, p = 0.07) (Add-itional file 6: Table S4). No statistically significant inter-actions were observed between IgG subclass levels andinhaled steroid therapy in both cohorts for either exacer-bations or hospitalizations.DiscussionIn this study, we found that IgG subclass deficiency wascommon in COPD, affecting approximately 1 in 5 patientswho were at increased risk for COPD exacerbations. Inaddition, progressively lower IgG1 and IgG2 levels wereassociated with increased risk of exacerbations and hospi-talizations in two different COPD cohorts. Moreover, theprotective effect of IgG1 and IgG2 was even more pro-nounced on hospitalizations compared to exacerbations inboth analyzed cohorts, emphasizing the potential role ofIgG subclass levels in modulating the frequency and sever-ity of COPD exacerbations.Table 4 Unadjusted and adjusted hazard ratios related tohospitalizations according to IgG subclass levels in the MACROand STATCOPE cohortsMACRO – Time to first HospitalizationIgGa Univariate analysis Multivariate analysisbHazard Ratio (95% CI) P Hazard Ratio (95% CI) PIgG1 1.51 (1.16–1.99) 0.003 1.52 (1.15–2.02) 0.004IgG2 1.37 (1.12–1.67) 0.002 1.33 (1.08–1.64) 0.007IgG3 1.03 (0.85–1.24) 0.79 1.01 (0.84–1.44) 0.90IgG4 1.01 (0.89–1.12) 0.99 1.00 (0.88–1.12) 0.94STATCOPE – Time to first HospitalizationIgGa Univariate analysis Multivariate analysisbHazard Ratio (95% CI) P Hazard Ratio (95% CI) PIgG1 1.33 (0.98–1.80) 0.07 1.27 (0.91–1.76) 0.16IgG2 1.39 (1.11–1.74) 0.004 1.43 (1.12–1.83) 0.004IgG3 0.97 (0.79–1.19) 0.75 0.90 (0.73–1.12) 0.36IgG4 1.08 (0.95–1.22) 0.22 1.06 (0.93–1.20) 0.39aIgG subclass levels expressed using a negative-log transformation (base 2), with one-unit increase on this log scale being equivalent to a 50% decrease of IgG subclasslevels. bAdjusted hazard ratios for IgG levels were calculated using a Cox ProportionalHazards model with adjustments for the following covariates: study drug (azithromycinvs. placebo in MACRO - First cohort; simvastatin vs. placebo in STATCOPE - Replicationcohort), age, gender, ethnicity (Caucasian vs. Non-Caucasian), FEV1 (% of predictedvalues), smoking status (current vs. former smokers), oxygen dependence, inhaledcorticosteroid use, and treatment with systemic steroids in previous year. Legend: 95%CI = 95% Confidence IntervalFig. 2 Correlation of IgG1 (upper panel) and IgG2 (lower panel) levels (log-transformed, base2) and the predicted risk of COPD exacerbations. Foreach participant, the predicted risks were calculated using logistic regression models with adjustments for the following covariates: study drug(azithromycin vs. placebo in MACRO - First cohort; simvastatin vs. placebo in STATCOPE - Replication cohort), age, gender, ethnicity (Caucasian vs.non-Caucasian), FEV1 (% of predicted values), smoking status (current vs. former smokers), oxygen dependence, inhaled corticosteroid use, andtreatment with systemic steroids in previous yearLeitao Filho et al. Respiratory Research  (2018) 19:30 Page 6 of 10Several prior reports have described the prevalence ofIgG subclass deficiency in COPD patients [15–17].O’Keeffe et al. observed an IgG subclass deficiency in 15of 58 COPD patients (25.9%) [15]. Qvarfordt et al. mea-sured IgG subclass levels in 33 smokers with chronicbronchitis and recurrent exacerbations, identifying anIgG subclass deficiency in 6 patients (18%) [16]. We ob-tained similar results, as at least one IgG subclass defi-ciency was detected in 17.7% and 20.4% of participantswith COPD in MACRO and STATCOPE cohorts, re-spectively. Concerning the distribution of IgG subclassdeficiency, we identified IgG3 as the most frequent(6.7% in MACRO and 9.0% of cases in STATCOPE),which has also been reported by other studies [16, 23].Our findings suggest that the frequency of IgG abnor-malities may be even higher among COPD patients whopresent with multiple exacerbations, as we detected ahigher frequency of IgG1 and IgG2 deficiency among re-current exacerbators. This is supported by a previousarticle, in which IgA, IgG, and IgM were measured in 42COPD patients who had 2 or more moderate to severeexacerbations per year [24]. The authors detected in 29patients (69%) evidence of an antibody deficiency syn-drome, with 8 patients receiving a diagnosis of commonvariable immunodeficiency syndrome, whose diagnosisrequires reduced total IgG levels [25]. Similarly, weobserved significantly lower IgG1 and IgG2 levels amongCOPD participants who developed exacerbations and re-quired hospitalizations in both cohorts.In our Cox proportional hazards regression models, re-ductions in IgG1 and IgG2 levels in the First cohort wereassociated with an increased risk of COPD exacerbationsand hospitalizations. These results remained statisticallysignificant after adjustments for differences in age, gender,race, study drug, smoking status, oxygen dependence, FEV1(% of predicted values), use of inhaled steroids and treat-ment with systemic steroids in the year prior to enrollment.Similar results were detected using a second independentcohort (thus increasing the validity of our findings) [26],with the exception that only IgG2 remained as an inde-pendent predictor of hospitalizations. Additionally, we alsoshowed that participants with IgG1 and/or IgG2 deficiencypresented with higher rates of COPD exacerbations perperson-year during follow-up. To the best of our know-ledge, this is the first report to demonstrate that lower IgGsubclass levels are independent risk factors for both COPDexacerbations (IgG1 and IgG2) and hospitalizations (IgG2).Our results are in line with some important concepts re-lated to the biology of IgG subclasses. IgG1 and IgG2 are themajor components of serum total IgG levels, explaining whypatients with deficiencies in either one of these subclasses(or both) are more likely to present with reduced total IgGFig. 3 Correlation of IgG1 (upper panel) and IgG2 (lower panel) levels (log-transformed, base2) and the predicted risk of hospitalizations. For eachparticipant, the predicted risks were calculated using logistic regression models with adjustments for the following covariates: study drug(azithromycin vs. placebo in MACRO - First cohort; simvastatin vs. placebo in STATCOPE - Replication cohort), age, gender, ethnicity (Caucasian vs.non-Caucasian), FEV1 (% of predicted values), smoking status (current vs. former smokers), oxygen dependence, inhaled corticosteroid use, andtreatment with systemic steroids in previous yearLeitao Filho et al. Respiratory Research  (2018) 19:30 Page 7 of 10levels (i.e., hypogammaglobulinemia) [10, 11, 25]. We ob-served that 97.3% and 74.2% of patients diagnosed with IgG1and IgG2 deficiencies (considering the merged cohort) alsomet criteria for hypogammaglobulinemia, respectively. Thus,patients with lower IgG1 and IgG2 levels, through a reduc-tion in total IgG levels, may present with an increased risk ofCOPD exacerbations and hospitalizations, as these outcomesare mostly triggered by respiratory infections [4]. This is con-sistent with a previous work of our group, in which COPDpatients with hypogammaglobulinemia showed a higher fre-quency of COPD exacerbations and hospitalizations com-pared to patients with normal total IgG levels [14].Additionally, patients with IgG2 deficiency usually presentwith impaired polysaccharide responses, leading to a highersusceptibility to infections caused by encapsulated pathogens(e.g., Streptococcus pneumoniae and Haemophilus influenzaetype B) [10, 12, 13], which are frequently implicated in bac-terial COPD exacerbations [4].It has been shown that not all patients with lower IgG sub-classes present with recurrent infections, with some healthyindividuals also being diagnosed with one or more IgG sub-classes deficiencies [11]. Conversely, there is a rationale forevaluating patients with recurrent infections for possible IgGsubclass deficiency. Kim JH et al. observed that one-third ofpatients with chronic airway diseases (asthma, COPD, orasthma COPD overlap syndrome) who had received a previ-ous diagnosis of IgG subclass deficiency (n= 59) presentedwith a history of recurrent respiratory infections [23]. Add-itionally, these patients showed a higher frequency of hospi-talizations and a faster FEV1 decline during follow-up. Ourresults indicate that IgG1 or IgG2 subclass deficiency maycontribute to a higher exacerbation risk, and thus we advo-cate that measurement of IgG subclass levels be consideredin any COPD patient with a history of several exacerbationsor previous hospitalizations. Moreover, in this scenario, evenif a COPD patient shows normal total IgG levels, it is stillpossible that an IgG1 or IgG2 subclass deficiency may bepresent, potentially contributing to an increased propensityto COPD exacerbations. In a previous article, Vendrell et al.found that 11% of patients with bronchiectasis of unknownetiology with normal total IgG levels (n= 107) still had anantibody production deficiency, with 50% of those presentingwith IgG2 deficiency [27].The management of IgG subclass deficiency includesvaccinations, prophylactic antibiotics, treatment of allergyand sinopulmonary disease (if present) and cautious useof immunoglobulin G replacement therapy - IVIG (for pa-tients who persist with recurrent infections despite theseinterventions) [25]. In one previous study, administrationof IVIG reduced the risk of respiratory infections by over50% in 92/132 adults with IgG subclass deficiency present-ing with recurrent infections [28]. In another study, IVIGsignificantly decreased the frequency of moderate to se-vere exacerbations by nearly 90% in 14 COPD patients(from an annual COPD exacerbation rate of 4.7 ± 3.1 to0.6 ± 1.0 per patient)[29]. Interestingly, 64.3% of the par-ticipants (9/14) had evidence of hypogammaglobulinemia.Our study had several limitations. Firstly, we were onlyable to measure IgG subclass levels at baseline, and thus wecannot determine how IgG subclass levels fluctuate over thecourse of COPD or with the implementation of variousCOPD treatments. Secondly, we only measured IgG subclasslevels once, despite standard laboratory recommendationsthat two measurements be performed at least 1 month apartto confirm deficiency [11, 25]. Thirdly, we did not investigatewhether patients with IgG subclass deficiency had impairedhumoral immunity, usually evaluated by measuring antibodylevels after a pneumococcal vaccination [11, 13, 25]. Finally,since this is a retrospective, observational study, we can onlydescribe an association between low serum IgG1 or IgG2levels and COPD exacerbations. Although IgG subclass defi-ciency may serve as a biomarker to identify subjects at in-creased risk of exacerbation, our study was not designed todetermine if low serum IgG subclass levels are linked on acausal pathway to COPD exacerbations. It is worth mention-ing we analyzed samples from moderate-to-severe COPDpatients who were at increased risk of COPD exacerbations;thus the effect of lower IgG levels on the outcomes of COPDpatients with mild disease or without a history of prior exac-erbations was not evaluated by our study.ConclusionsOur data support that IgG subclass deficiency is relativelycommon among COPD patients and that decreased IgG1and IgG2 levels result in an increased risk of adverse out-comes in COPD. Future prospective studies are needed tobetter elucidate the impact of these IgG subclass deficien-cies on the management of high-risk COPD patients andwhether they represent a modifiable risk factor for COPDexacerbations and hospitalizations through IVIG.Additional filesAdditional file 1: Table S1. Prevalence of IgG subclass deficiencies inMACRO and STATCOPE cohorts. (DOCX 15 kb)Additional file 2: Table S2. Median and interquartile range related toeach IgG subclass according to the presence or absence of correspondingIgG subclass deficiency in MACRO and STATCOPE cohorts. (DOCX 15 kb)Additional file 3: Table S3. Median and interquartile range related toeach IgG subclass according to the presence or absence ofcorresponding IgG subclass deficiency in the merged dataset (MACROand STATCOPE cohorts combined). (DOCX 15 kb)Additional file 4: Figure S1. Comparison of IgG subclass levelsaccording to exacerbation status in MACRO – First cohort (left panel) andSTATCOPE – Replication cohort (right panel). Error bars represent 95%confidence interval. (DOCX 205 kb)Additional file 5: Figure S2. Comparison of IgG subclass levelsaccording to hospitalization status in MACRO – First cohort (left panel)and STATCOPE – Replication cohort (right panel) cohorts. Error barsrepresent 95% confidence interval. (DOCX 243 kb)Leitao Filho et al. Respiratory Research  (2018) 19:30 Page 8 of 10Additional file 6: Table S4. Interactions between IgG subclass levelsand inhaled steroid use at enrollment and use of systemic steroids inprevious 12 months for both outcomes of interest (time to firstexacerbation and time to first hospitalization). (DOCX 15 kb)AbbreviationsCI: Confidence intervals; COPD: Chronic obstructive pulmonary disease;FEV1: Forced expiratory volume in 1 s; HR: Hazard ratio; Ig: Immunoglobulins;IgG: Immunoglobulin G; IVIG: Immunoglobulin G replacement therapy;LLN: Lower limit of normal; MACRO: Macrolide Azithromycin for Preventionof Exacerbations of COPD trial; STATCOPE: Simvastatin for the Prevention ofExacerbations in Moderate-to-Severe COPD trialAcknowledgmentsWe thank the COPD Clinical Research Network for conducting both the MACROand STATCOPE trials and for providing us access to all available samples in bothcohorts. Dr. Leitao Filho FS gratefully acknowledges postdoctoral support fromCNPq-Brazil and the Canadian Institutes of Health Research (CIHR) Integratedand Mentored Pulmonary and Cardiovascular Training program (IMPACT).FundingThis work was supported by the BC Lung Association and the CanadianRespiratory Research Network (CRRN). The MACRO and STATCOPE trials werefunded by the US National Heart Blood and Lung Institute (NHLBI).Availability of data and materialsThe datasets used and/or analyzed during the current study are availablefrom the corresponding author on reasonable request.Authors’ contributionsFSLF, SWR, RMR, SDA, RSS and DDS: conception and design of the work;RMR, SDA, AM, RA, GJC, PGW, JEC, SCL, MKH, FJM, and DDS: Acquisition ofdata; All authors: analysis and interpretation of data; FSLF, SWR, SDA, RMR,RSS and DDS: Drafting the manuscript; All authors: Review and approval ofthe final version of the manuscript for important intellectual content.Ethics approval and consent to participateBoth trials received approval from each participating institutions’ local research ethicsboard. This analysis received approval from the University of British Columbia/Providence Health Care Research Ethics Committee (Approval No. H16–03232).Consent for publicationNot applicable.Competing interestsThe authors declare that they have no competing interests.Publisher’s NoteSpringer Nature remains neutral with regard to jurisdictional claims inpublished maps and institutional affiliations.Author details1Centre for Heart Lung Innovation, St. Paul’s Hospital, Vancouver, BC V6Z 1Y6,Canada. 2Department of Medicine (Division of Respiratory Medicine),University of British Columbia, Vancouver, BC, Canada. 3Federal University ofCeará, Fortaleza, Ceará, Brazil. 4University of Ulsan College of Medicine, UlsanUniversity Hospital, Ulsan, South Korea. 5Department of Pathology andLaboratory Medicine, University of British Columbia, Vancouver, BC, Canada.6Department of Thoracic Medicine and Surgery, Lewis Katz school ofMedicine at Temple University, Philadelphia, PA, USA. 7Department ofMedicine, University of California San Francisco, San Francisco, CA, USA.8Pulmonary Sciences and Critical Care Medicine, University of Colorado,Denver, CO, USA. 9School of Public Health, University of Minnesota,Minneapolis, MN, USA. 10Department of Internal Medicine, University ofMichigan, Ann Arbor, MI, USA. 11Joan and Sanford I. Weill Department ofMedicine, Weill Cornell Medical College, Cornell University, New York, NY,USA. 12Department of Medicine, University of Ottawa, Ottawa, ON, Canada.13Division of Pulmonary and Critical Care Medicine, University of MarylandSchool of Medicine, Baltimore, MD, USA.Received: 15 January 2018 Accepted: 31 January 2018References1. Vogelmeier CF, Criner GJ, Martinez FJ, Anzueto A, Barnes PJ, Bourbeau J, etal. Global Strategy for the Diagnosis, Management, and Prevention ofChronic Obstructive Lung Disease 2017 Report. GOLD Executive Summary.Am. J. Respir. Crit. Care Med. 2017;195(5):557–82.2. Anzueto A. Impact of exacerbations on COPD. Eur. Respir. Rev. 2010;19(116):113–8.3. Soler-Cataluna JJ, Martinez-Garcia MA, Roman Sanchez P, Salcedo E, NavarroM, Ochando R. Severe acute exacerbations and mortality in patients withchronic obstructive pulmonary disease. Thorax. 2005;60(11):925–31.4. Miravitlles M, Anzueto A. Role of infection in exacerbations of chronicobstructive pulmonary disease. Curr. Opin. Pulm. Med. 2015;21(3):278–83.5. Umetsu DT, Ambrosino DM, Quinti I, Siber GR, Geha RS. Recurrentsinopulmonary infection and impaired antibody response to bacterialcapsular polysaccharide antigen in children with selective IgG-subclassdeficiency. N. Engl. J. Med. 1985;313(20):1247–51.6. Aucouturier P, Lacombe C, Bremard C, Lebranchu Y, Seligmann M, GriscelliC, et al. Serum IgG subclass levels in patients with primaryimmunodeficiency syndromes or abnormal susceptibility to infections. Clin.Immunol. Immunopathol. 1989;51(1):22–37.7. Popa V, Kim K, Heiner DC. IgG deficiency in adults with recurrent respiratoryinfections. Ann. Allergy. 1993;70(5):418–24.8. Popa V. Airway obstruction in adults with recurrent respiratory infectionsand IgG deficiency. Chest. 1994;105(4):1066–72.9. Sethi S, Murphy TF. Infection in the pathogenesis and course of chronicobstructive pulmonary disease. N. Engl. J. Med. 2008;359(22):2355–65.10. Schroeder HW Jr, Cavacini L. Structure and function of immunoglobulins. J.Allergy Clin. Immunol. 2010;125(2 Suppl 2):S41–52.11. Agarwal S, Cunningham-Rundles C. Assessment and clinical interpretationof reduced IgG values. Ann. Allergy Asthma Immunol. 2007;99(3):281–3.12. Vidarsson G, Dekkers G, Rispens T. IgG subclasses and allotypes: fromstructure to effector functions. Front Immunol. 2014;5:520.13. Maguire GA, Kumararatne DS, Joyce HJ. Are there any clinicalindications for measuring IgG subclasses? Ann. Clin Biochem.2002;39(Pt 4):374–7.14. Leitao Filho FS, Won Ra S, Mattman A, Schellenberg RS, Fishbane N, Criner GJ,et al. Serum IgG and risk of exacerbations and hospitalizations in chronicobstructive pulmonary disease. J. Allergy Clin. Immunol. 2017;140(4):1164–7. e615. O’Keeffe S, Gzel A, Drury R, Cullina M, Greally J, Finnegan P.Immunoglobulin G subclasses and spirometry in patients with chronicobstructive pulmonary disease. Eur. Respir. J. 1991;4(8):932–6.16. Qvarfordt I, Riise GC, Andersson BA, Larsson S. IgG subclasses in smokers withchronic bronchitis and recurrent exacerbations. Thorax. 2001;56(6):445–9.17. Karnak D, Beder S, Kayacan O, Karaca L. IgG Subclass Levels in ELF in COPD.Turk J Med Sci. 2001;31:235–41.18. Zhou A, Zhou Z, Zhao Y, Chen P. The recent advances of phenotypes inacute exacerbations of COPD. Int J Chron Obstruct Pulmon Dis. 2017;12:1009–18.19. Albert RK, Connett J, Bailey WC, Casaburi R, Cooper JA Jr, Criner GJ, et al.Azithromycin for prevention of exacerbations of COPD. N. Engl. J. Med.2011;365(8):689–98.20. Criner GJ, Connett JE, Aaron SD, Albert RK, Bailey WC, Casaburi R, et al.Simvastatin for the prevention of exacerbations in moderate-to-severeCOPD. N. Engl. J. Med. 2014;370(23):2201–10.21. Schauer U, Stemberg F, Rieger CH, Borte M, Schubert S, Riedel F, et al. IgGsubclass concentrations in certified reference material 470 and referencevalues for children and adults determined with the binding site reagents.Clin. Chem. 2003;49(11):1924–9.22. Stiehm ER, Ochs HD, Winkelstein JA. Immunologic Disorders in Infants andChildren. 5th ed. Philadelphia: Saunders; 2004. p. 1512.23. Kim JH, Park S, Hwang YI, Jang SH, Jung KS, Sim YS, et al. ImmunoglobulinG Subclass Deficiencies in Adult Patients with Chronic Airway Diseases. J.Korean Med. Sci. 2016;31(10):1560–5.24. McCullagh BN, Comellas AP, Ballas ZK, Newell JD Jr, Zimmerman MB, AzarAE. Antibody deficiency in patients with frequent exacerbations of ChronicObstructive Pulmonary Disease (COPD). PloS One. 2017;12(2):e0172437.25. Bonilla FA, Khan DA, Ballas ZK, Chinen J, Frank MM, Hsu JT, et al. Practiceparameter for the diagnosis and management of primaryimmunodeficiency. J. Allergy Clin. Immunol. 2015;136(5):1186–205 e1–78.Leitao Filho et al. Respiratory Research  (2018) 19:30 Page 9 of 1026. Carlson MD, Morrison RS. Study design, precision, and validity inobservational studies. J. Palliat. Med. 2009;12(1):77–82.27. Vendrell M, de Gracia J, Rodrigo MJ, Cruz MJ, Alvarez A, Garcia M, et al.Antibody production deficiency with normal IgG levels in bronchiectasis ofunknown etiology. Chest. 2005;127(1):197–204.28. Olinder-Nielsen AM, Granert C, Forsberg P, Friman V, Vietorisz A, BjorkanderJ. Immunoglobulin prophylaxis in 350 adults with IgG subclass deficiencyand recurrent respiratory tract infections: a long-term follow-up. Scand. J.Infect. Dis. 2007;39(1):44–50.29. Cowan J, Gaudet L, Mulpuru S, Corrales-Medina V, Hawken S, Cameron C, etal. A Retrospective Longitudinal Within-Subject Risk Interval Analysis ofImmunoglobulin Treatment for Recurrent Acute Exacerbation of ChronicObstructive Pulmonary Disease. PloS One. 2015;10(11):e0142205.•  We accept pre-submission inquiries •  Our selector tool helps you to find the most relevant journal•  We provide round the clock customer support •  Convenient online submission•  Thorough peer review•  Inclusion in PubMed and all major indexing services •  Maximum visibility for your researchSubmit your manuscript atwww.biomedcentral.com/submitSubmit your next manuscript to BioMed Central and we will help you at every step:Leitao Filho et al. Respiratory Research  (2018) 19:30 Page 10 of 10


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items