Duration of frequent or severe respiratory tract infection in adults before diagnosis of IgG subclass deficiency

Many adults with IgG subclass deficiency (IgGSD) experience long intervals of frequent/severe respiratory tract infection before IgGSD diagnosis, but reasons for delays in IgGSD diagnoses are incompletely understood. We performed a retrospective study of 300 white adults (ages ≥18 y) with IgGSD including frequency analyses of age at IgGSD diagnosis, duration of frequent/severe respiratory tract infection before IgGSD diagnosis, and age at onset of frequent/severe infection (calculated). We performed multivariable regressions on age at diagnosis, infection duration, and age at infection onset using these variables, as appropriate: sex; age at diagnosis; diabetes; autoimmune condition(s); atopy; allergy; corticosteroid use; body mass index; serum immunoglobulin isotype levels; blood lymphocyte subsets; three IgGSD-associated human leukocyte antigen-A and -B haplotypes; and referring physician specialties. Mean age at diagnosis was 50 ± 12 (standard deviation) y (median 50 y (range 19–79)). There were 247 women (82.3%). Mean infection duration at IgGSD diagnosis was 12 ± 13 y (median 7 y (range 1–66)). Mean age at infection onset was 38 ± 16 y (median 38 y (range 4, 76)). Age at infection onset was ≥18 y in 95.7% of subjects. Regressions on age at diagnosis and infection duration revealed no significant associations. Regression on age at infection onset revealed one positive association: age at diagnosis (p <0.0001). We conclude that the median duration of frequent/severe respiratory tract infection in adults before IgGSD diagnosis was 7 y. Older adults may be diagnosed to have IgGSD after longer intervals of infection than younger adults. Duration of frequent/severe respiratory tract infection before IgGSD diagnosis was not significantly associated with routine clinical and laboratory variables, including referring physician specialties.


Introduction
Immunoglobulin (Ig) G subclass deficiency (IgGSD) in adults is characterized by frequent or severe respiratory tract infection, suboptimal IgG response to polyvalent pneumococcal a1111111111 a1111111111 a1111111111 a1111111111 a1111111111 polysaccharide vaccination, female predominance, and increased prevalence of autoimmune conditions [1][2][3]. Many adults with IgGSD experience long intervals of frequent or severe respiratory tract infection before IgGSD diagnosis [4][5][6]. Delays in diagnosis of IgGSD in adults have been associated with greater morbidity due to major infections [6][7][8]. We postulated that analyses of age at IgGSD diagnosis, duration of frequent or severe respiratory tract infection before IgGSD diagnosis, and age at onset of frequent or severe respiratory tract infection could provide insights into the chronology of IgGSD manifestations and delays of IgGSD diagnosis in adults.
To learn more, we performed a retrospective study of 300 unrelated non-Hispanic whites diagnosed to have IgGSD as adults (ages �18 y). We analyzed distributions of ages at IgGSD diagnosis, duration of frequent or severe respiratory tract infection before IgGSD diagnosis, and ages at onset of frequent or severe respiratory tract infection. We also performed multivariable regressions on age at IgGSD diagnosis, duration of frequent or severe respiratory tract infection, and age at onset of frequent or severe respiratory tract infection using these independent routine clinical and laboratory variables, as appropriate: sex; age at diagnosis; diabetes; autoimmune condition(s); atopy; allergy; corticosteroid use; body mass index; serum Ig isotype levels; blood lymphocyte subsets; three human leukocyte antigen (HLA)-A and -B haplotypes associated with IgGSD in adults; and specialties of referring physicians. Herein, we determined that the median duration of frequent/severe respiratory tract in adults before IgGSD diagnosis was 7 y and that the duration of frequent/severe respiratory tract infection before IgGSD diagnosis was not significantly associated with routine clinical and laboratory variables, including referring physician specialties. We discuss our findings in the context of previous reports of IgGSD diagnosis in adults.

Ethics statement
This work was performed according to the principles of the Declaration of Helsinki [9]. Western Institutional Review Board provided an exemption under 45 CFR 46.101(b)(4) pertinent to this study on 18 October 2018 (submission 2535878-44170911; 2 October 2018). Obtaining informed consent was not required because this study involved retrospective chart review and analyses of observations recorded in routine medical care.

Patient selection
We studied consecutive unrelated self-identified non-Hispanic white adults (ages �18 y) referred to a single outpatient referral practice because they had frequent or severe upper or lower respiratory tract infection and were diagnosed to have IgGSD [1-3] before 2 October 2018. Subnormal IgG4 level alone was not a criterion for IgGSD diagnosis [1][2][3]. We compiled referring physician specialties. Upper respiratory tract infection was defined as reports of sinusitis, otitis media, mastoiditis, pharyngitis, and tonsillitis. Lower respiratory tract infection was defined as reports of bronchitis, pneumonia, and bronchiectasis. Frequent infection was defined as four or more episodes per year that required antibiotic therapy. Severe infection was defined as any infection that required in-hospital treatment.
We excluded adults with Ig deficiency other than IgGSD [3], including: common variable immunodeficiency, ataxia-telangiectasia, X-linked agammaglobulinemia; hypogammaglobulinemia associated with acute infection, monoclonal gammopathy, B-cell neoplasms or other malignancies, organ transplantation, immunosuppressive therapy, anti-cancer chemotherapy, carbamazepine therapy, or increased Ig loss; infection with parasites, Mycobacterium sp., or human immunodeficiency virus; and incomplete evaluations.

Duration of infection and age at onset of infection
We compiled reports of duration of frequent or severe respiratory tract infection before IgGSD diagnosis. Age at onset of frequent or severe infection was defined as the difference between age at IgGSD diagnosis and reported duration of frequent or severe respiratory tract infection before IgGSD diagnosis.

Other conditions
We classified diabetes according to the criteria of the American Diabetes Association [10]. Autoimmune condition(s) and atopy (allergic asthma, allergic rhinitis, and allergic dermatitis/ eczema) were diagnosed by referring physicians. Other allergy included urticaria, angioedema, or anaphylaxis [2]. We defined a dichotomous corticosteroid therapy variable as previously described [2]. Body mass index was computed as kg/m 2 .
HLA-A and -B types and haplotypes were detected using low-resolution DNA-based typing (polymerase chain reaction/sequence-specific oligonucleotide probe) [12]. Control subjects were 751 unrelated white subjects from Alabama who underwent HLA-A and -B haplotyping to establish paternity [13]. For analyses, we selected the three most common haplotypes that occur with greater prevalence in Alabama non-Hispanic white adults with IgGSD than in control subjects (A � 01,B � 08; A � 02,B � 44; and A � 03,B � 07) [2,12].

Statistics
The dataset for analyses consisted of complete observations on 300 adults. All data underlying the findings reported in this work are provided in a Supporting Information file (S1 File). All age at diagnosis, infection duration, and age at infection onset data were expressed as the nearest whole year. IgG4 levels <1 mg/dL were imputed as 0.5 mg/dL. Descriptive data are displayed as enumerations, percentages, mean ± 1 SD, or median (range). We evaluated continuous data for normality using d'Agostino's test and normal probability plots. We performed backward stepwise regressions on age at IgGSD diagnosis, duration of frequent or severe infection before IgGSD diagnosis, and age at onset of frequent or severe infection using these independent routine clinical and laboratory variables, as appropriate: sex; age at diagnosis; diabetes; autoimmune condition(s); atopy; allergy; corticosteroid use; body mass index; serum Ig isotype levels; blood lymphocyte subsets; HLA-A � 01,B � 08; -A � 02,B � 44; and -A � 03, B � 07 haplotype positivity (heterozygosity or homozygosity); and referring physician specialties. We defined values of p <0.05 to be significant. Analyses were performed with Excel 2000 (Microsoft Corp., Redmond, WA, USA) and GB-Stat (v. 10.0, 2003, Dynamic Microsystems, Inc., Silver Spring, MD, USA).

Duration of infection before IgGSD diagnosis
These data were not normally distributed (Fig 2). Mean duration of frequent or severe respiratory tract infection before IgGSD diagnosis was 12 ± 13 y (median 7 y (range 1-66)). Proportions of adults who reported that they experienced frequent or severe respiratory tract infection �1 y and �2 y before IgGSD diagnosis were 15.0% and 28.0%, respectively. Sixtyfour percent of adults experienced frequent or severe respiratory tract infection for �10 y before IgGSD diagnosis. Nine adults (3.0%) experienced frequent or severe respiratory tract infection for >40 y before IgGSD diagnosis.

Age at onset of infection
Age at onset of frequent or severe respiratory tract infection data were not normally distributed (Fig 3). Mean age at onset of infection was 38 ± 16 y (median 38 y (range 4, 76)). One- hundred and forty-one adults (47.0%) reported that they experienced frequent or severe respiratory infection at ages �35 y. Thirteen adults (4.3%; 2 men, 11 women) reported that they experienced frequent or severe respiratory infection before age 18 y (range 4-17 y). The remaining 287 (95.7%) adults reported that they experienced frequent or severe respiratory infection at age �18 y.

Regression on age at IgGSD diagnosis
We performed a backward stepwise variable regression on age at IgGSD diagnosis using these independent variables: sex; diabetes; autoimmune condition(s); atopy; allergy; corticosteroid use; body mass index; serum Ig isotype levels; blood lymphocyte subsets; three HLA-A and -B haplotypes; and referring physician specialties. This regression revealed no significant associations.

Regression on duration of infection before IgGSD diagnosis
We performed a backward stepwise variable regression on duration of infection before IgGSD diagnosis using these independent variables: sex; age at diagnosis; diabetes; autoimmune condition(s); atopy; allergy; corticosteroid use; body mass index; serum Ig isotype levels; blood lymphocyte subsets; three HLA-A and -B haplotypes; and referring physician specialties. This regression revealed no significant associations.

Regression on age at onset of infection
We performed a backward stepwise variable regression on age at onset of frequent or severe respiratory tract infection using these independent variables: sex; age at diagnosis; diabetes; autoimmune condition(s); atopy; allergy; corticosteroid use; body mass index; serum Ig isotype levels; blood lymphocyte subsets; three HLA-A and -B haplotypes; and referring physician specialties. This regression revealed a single positive association: age at diagnosis (p <0.0001). This regression accounted for 42.2% of the deviance of age at onset of infection (ANOVA p of regression <0.0001).

Discussion
The mean age (range) at IgGSD diagnosis in the present adults (50 y ) is consistent with those of three other adult IgGSD cohorts from The Netherlands and US: 49 y (33-70) [14]; 51 y (18-89) [12]; and 47 y (17-71) [15], respectively. In 59 Korean adults with either asthma or chronic obstructive pulmonary disease, the mean age at subsequent IgGSD diagnosis was 61 ± 16 y [6]. Our age at IgGSD diagnosis data were normally distributed, consistent with a previous report that included 398 Alabama adults with IgGSD [2]. In other studies, age at IgGSD diagnosis in adults classified by IgG subclass immunophenotypes was not significantly associated with routine clinical and laboratory variables [16,17]. Because the present regression on age at IgGSD diagnosis also revealed no significant associations with the same clinical and laboratory variables, it is likely that heritable or acquired factors or health care delivery features other than those we studied influence age at IgGSD diagnosis in adults. Duration of frequent or severe respiratory tract infection before IgGSD diagnosis in the present adults was great (mean 12 y, median 7 y). In a cohort of English adults, the median duration of frequent or severe respiratory tract infection before diagnosis of IgGSD or specific antibody deficiency was 11 y [5]. Most Korean adults with asthma or chronic obstructive pulmonary disease were diagnosed to have IgGSD more than 5 y after initial treatment of their respective pulmonary conditions [6]. These and other reports [4,7] substantiate that many adults with IgGSD experience long intervals of frequent or severe respiratory tract infection before IgGSD diagnosis.
Frequent or severe respiratory tract infection is the most common manifestation of undiagnosed IgGSD [4,[18][19][20]. By definition, all of the present adults had numerous out-patient or in-hospital physician encounters for management of respiratory tract infection before referral and IgGSD diagnosis, but multivariable regression on duration of infection revealed no significant associations with routine clinical and laboratory variables. It is plausible that frequent or severe respiratory tract infection in some adults with undiagnosed IgGSD was interpreted as a sequel of diabetes [21,22], chronic sinusitis or rhinosinusitis [19,23], autoimmune condition(s) [24,25], asthma [26][27][28], allergic rhinitis [29], or chronic obstructive pulmonary disease [4,6,18]. Nonetheless, these observations also suggest that lack of knowledge about IgGSD among physicians who treat adults with respiratory tract infection may inadvertently increase the duration of frequent or severe respiratory tract infection before IgGSD diagnosis.
Diabetes was diagnosed in more than 10% of the present adults with IgGSD, a proportion similar to that of adults ages �18 y with diabetes in the general Alabama population (12%) [30]. Diabetes was not significantly associated with the age of onset or duration of frequent or severe respiratory tract infection in the present cohort, although this does not exclude a role of diabetes in antigen response, immunoglobulin synthesis, or infection susceptibility in adults with or without IgGSD. In a prospective study of 247 adults with diabetes, 48 adults had either a history of infections or a globulin level <2.2 g/dL [22]. The prevalence of subnormal IgG, IgA, or IgM, alone or in combination, was significantly higher in the adults with diabetes than in the general population and could explain in part the greater susceptibility of adults with diabetes to infection [22].
Physicians of different specialties treat patients with frequent or severe respiratory tract infection. Otolaryngologists, primary care physicians, rheumatologists, and pulmonologists referred 92.3% of the present adults, consistent with results of a previous study of adults with IgGSD or common variable immunodeficiency [2]. In the present study, duration of frequent or severe respiratory tract infection or age at onset of frequent or severe respiratory tract infection was not significantly associated with referring physician specialty. In another study, allergists and clinical immunologists, otorhinolaryngologists, and pulmonologists in aggregate referred 55% and primary care physicians referred 17% of 244 Canadian adults with immune deficiency diagnosed in a university center for adults with primary immune deficiency [31]. Pre-referral measurement of serum Ig levels was performed in only one-third of the 244 Canadian adults [31]. In a 1989 report from northwest England, the median delay in diagnosis of primary antibody deficiency in adults was 5.5 y [7]. After introduction of UK national diagnosis and treatment guidelines in 1995, median diagnostic delay in adults and children with primary antibody deficiency decreased to 2 y by 2005, although the proportion of referrals from primary care physicians remained low [8]. These observations suggest that educating physicians of many specialties about clinical and laboratory characteristics of immune deficiency could reduce delays in IgGSD diagnosis in adults.
Ninety-six percent of the present adults reported that they experienced frequent or severe respiratory tract infection at ages �18 y, suggesting that IgGSD in most adults becomes clinically manifest in adulthood. Almost half of the present adults reported that they first experienced frequent or severe respiratory tract infection at ages �35 y. Our multivariable regression on age at onset of infection before IgGSD diagnosis revealed a significant positive association with age at IgGSD diagnosis only. This suggests that older adults are diagnosed to have IgGSD after longer intervals of frequent or severe respiratory tract infection than younger adults, after adjustment for other variables.
The principal significance of this study is that many of the present adults with frequent or severe respiratory tract infection experienced a delay of many months or years before the underlying cause of their infections was diagnosed, yet most of them had repeated healthcare encounters for treatment of respiratory tract infection before they were evaluated with Ig measures [32]. Consequently, effective infection prophylaxis with IgG [33] and other management were delayed, thus increasing infection-associated morbidity and expense.
A strength of this study is that our sample size provides statistical power to detect significant independent associations with age at IgGSD diagnosis, duration of frequent or severe respiratory tract infection before IgGSD diagnosis, and age at onset of frequent or severe respiratory tract infection in multivariable models. Limitations of this study include lack of serum Ig levels measured in the present adults long before IgGSD diagnosis and possible inaccuracy of reports of duration of frequent or severe respiratory tract infection before IgGSD diagnosis. Another limitation is the lack of observations on a) ostensibly healthy, age-and sex-matched control subjects from the general non-Hispanic white population and b) other control subjects who had frequent or severe respiratory tract infection without evidence of IgGSD or other antibody deficiency. Evaluation of other heritable or acquired factors or health care delivery features that may influence duration of frequent or severe respiratory tract infection in adults before IgGSD diagnosis and morbidity associated with delayed diagnosis of IgGSD was beyond the scope of this study.

Conclusions
Median duration of frequent or severe respiratory tract infection in adults before IgGSD diagnosis was 7 y. Older adults may be diagnosed to have IgGSD after longer intervals of infection than younger adults. Duration of frequent or severe respiratory tract infection before IgGSD diagnosis was not significantly associated with routine clinical and laboratory variables, including referring physician specialties.