Immunoglobulin subclasses and HLA alleles in immunoglobulin A deficiency

Objective : The term “IgA Deficiency (IgAD)” should be reserved for the individuals who do not have detectable disorders known to be associated with low IgA levels. IgG subclass deficiency or a lack of the IgG2 subclass that is specific against polysaccharide antigens, can be seen in many cases.Methods : Forty-five patients (27 males and 18 females; mean age 8.6 years, range 6.3 to 12.8 years) with IgA deficiency who had been admitted to the Department of Pediatric Immunology in Uludag University School of Medicine, Turkey, were included in this study. Serum immunoglobulin (Ig) class and IgG subclass levels, and HLA haplotypes were prospectively determined in patients and healthy controls.Results : Of the 45 patients with IgAD, 1 was found to have a low level of IgG in the serum. Serum Ig levels were also examined in the families of 22 patients. Five patients had low-normal levels of IgM, whilst one had low levels of IgA and IgG. The levels of IgG sublasses were assessed in 23 patients. One patient had a low level of IgG1 ; 2 had low levels of both IgG2 and IgG3, and 11 had low levels of IgG3. IgG subclass concentrations were found to be normal in control groups. HLA alleles were tested in 25 patients. An increased prevelence of HLA-A1, -B8, -B14, -DR1, -DR3, and -DR7 were previously observed in patients with Ig A deficiency. In this study, HLA-A1 allel was found in 3 patients (12 %), HLA-B14 in 3 patients (12%), HLA-DR1 in 10 patients (40 %), HLA-DR7 in 4 patients (16 %) and HLA-DR3 in 1 patient (4 %). HLA-B8 allel was not found in any patient. Twenty-five children with normal IgA levels have chosen as a control group. They had HLA-DR1 (36%), HLA-DR7 (16 %), HLA-B8 (8%), HLA-DR3 (16%). HLA-A1 was not found in any member of our control group.Conclusion : No statistically significant difference in HLA susceptibility alleles was found between patients and healthy controls. Our data suggest that there may be heterogenous HLA distribution patterns in IgA deficiency, or that HLA allel-associated tendency to IgA deficiency may be polygenic.     

IgG subclass deficiency in children with IgA deficiency presenting with recurrent or severe respiratory infections

A group of 22 children presenting with recurrent or severe respiratory tract infections who had low IgA levels (more than 2 SD below the mean for age) were examined for IgG subclass deficiency. Patients were screened for possible defects in neutrophil chemotaxis, bactericidal, fungicidal, and quantitative iodination activity, as well as for complement function. The majority of the patients showed IgG subclass levels below the mean for age. Nine of the children showed definite IgG subclass deficiency and at least two showed definite deficiency of more than one IgG subclass. The predominant subclass deficiency was found to be IgG1. While nine children showed IgG4 levels below the level detectable by the technique used, it is not possible to assess whether these patients are deficient in this isotype since some healthy subjects also give values below the level of detection. Most of the patients who had very low (1-6 mg/dl) or undetectable (less than 1 mg/dl) levels of serum IgA did not show IgG subclass deficiencies, while IgG subclass deficiencies were common among those with borderline low IgA levels (slightly more than 2 SD below the mean for age). Nine children showed total IgG levels close to 2 SD below mean for age, and at least six of these showed IgG subclass deficiency. The result suggests that patients with recurrent and/or severe respiratory infections who have borderline IgA and IgG levels may have IgG subclass deficiencies and if they do could benefit from immunoglobulin therapy.     

Immunoglobulin G subclass concentrations and infections in children and adolescents with severe asthma

It was the aim of this study to investigate possible dysfunctions of the humoral immune system in asthmatic children with frequent respiratory infections. Forty‐one severe asthmatics (7–15 years of age), classified according to the Second Brazilian Consensus in Asthma (1998), were divided into two groups: group I (n = 12) had recurrent respiratory infections; and group II (n = 29) were without recurrent respiratory infections. Immunoglobulin (Ig)G, IgA and IgM levels (nephelometry), and IgE (radioimmunoassay) and IgG subclasses (enzyme‐linked immunosorbent assay), were evaluated using standard methods. Asthmatics with recurrent infections presented with worse clinical evolution, an increased number of hospital admissions, and a higher need of medication than the children without recurrent infections. There were no significant differences between the mean values of IgG, IgA or IgM levels, or IgE or IgG subclasses, in patients of both groups. A complete IgA deficiency was detected in two patients of group I (one was associated with IgG subclass deficiency). Deficiency of one or more IgG subclasses was verified in eight of 12 (66%) children from group I and in 16/29 (55%) from group II. The following deficiencies were found in both groups: IgG₃ (10/41), IgG₄ (three of 41), IgG₂ (two of 41), IgG₁ (one of 41), IgG₃‐IgG₄ (four of 41), IgG₁‐IgG₃ (two of 41), and IgG₁‐IgG₃‐IgG₄ (one of 41). There were a higher proportion of children with low IgG₄ levels in group I than in group II (p = 0.01). To conclude, IgA and IgG subclass deficiencies were detected in both severely asthmatic groups, with a predominance of IgG₃ subclass deficiency. However, low IgG subclass levels appear not to be a suitable predictor of the development of infections in asthmatic children.     

IgG subclass deficiency in childhood. Changes in the antigen specific immune response and significance of low IgG4 level

The definition of IgG subclass deficiency and the correlations between low IgG subclass serum concentrations and high incidence of infections in certain patients are still obscure. Therefore 260 children from 6 months to 18 years with severe recurrent infections or a known immunodeficiency were screened for IgG subclass deficiency. Nine patients with severe IgG2 deficiency (Ig2 less than 0.3 g/l) and 35 patients with non-detectable IgG4 in immunoprecipitation were detected. One of these patients had a concomitant IgA deficiency, eight revealed an additional IgA and IgG2 deficiency, two an IgA, IgG2, and IgG3 deficiency, eighteen an IgG2 deficiency and one patient an IgG2 and IgG3 deficiency. The proportion of patients with non-detectable IgG4 in immunoprecipitation was 13.5% and thus in the same order of magnitude as described in the literature for healthy people. Our data show that there is no relation between low IgG4 serum levels and the increased occurrence of severe infections. In all patients investigated with non-detectable IgG4 in immunoprecipitation the gene for the heavy chain gene constant domain C gamma 4 could be detected by Southern blotting. Using a sensitive ELISA method IgG4 could be directly demonstrated in all patients at a serum level of 0.5-29 micrograms/ml. Specific IgG4 antibodies against protein antigens could not be detected in IgG4-deficient patients. Nevertheless total IgG antibodies against diphtheria and tetanus toxoid reached protecting titers. Patients with IgG2 deficiency showed an impaired immune response against polysaccharides from pneumococci and haemophilus influenzae type b.     

Immunoglobulin subgroups in children with febrile seizures

BACKGROUND: The aim of the present study was to determine whether or not there was a role for immunoglobulin (Ig) or IgG subgroups in the pathogenesis of febrile seizures (FS). METHODS: Serum levels of IgA, total IgG, IgM, IgE, IgGI, IgG2, IgG3 and IgG4 were measured in 34 children with FS and in 37 healthy children used as a control group. Both patients and controls were divided into two groups according to age (group I, 6-24 months; group II, 25-72 months). RESULTS: Compared with controls, mean IgG4 levels in patients were found to be decreased in both groups I and II (group I: 95 +/- 14 vs 57 +/- 5, respectively, P = 0.01; group II: 178.5 +/- 38.5 vs 65.1 +/- 24.5, respectively, P < 0.01), while mean IgG2 levels were found to be decreased only in group II patients (170 +/- 16 vs 103 +/- 22; P < 0.05). CONCLUSIONS: The results of the present study suggest that Ig subclass deficiencies may be responsible for the infections connected with FS or that they may be related to the pathogenesis of FS in some children.     

Clinico-immunologic aspects of IgG subclass deficiency

Serum IgG contains 4 subclasses, IgG1 (60-66%), IgG2 (20-30%), IgG3 (less than or equal to 5%) and IgG4. Individual subclasses vary with respect to their physicochemical and biological properties. IgG subclass concentrations in serum are age dependent. IgG1 and IgG3 reach near to adult levels around the age of 3, IgG2 and IgG4 after the age of 6. Antibodies of certain specificities generally belong to a certain isotype (subclass) due to the isotype restriction. Patients with subclass deficiencies often suffer from recurrent infections. Those with IgG2 deficiency (often occurring with IgA and IgG4 deficiency) develop recurrent infection of the upper and lower respiratory tract often caused by pyogenic microorganisms (Haemophilus, Pneumococcus). Since early initiation of IVIG substitution therapy has a beneficial effect on long term prognosis the importance of early diagnosis is apparent.     

Clinico-immunologic aspects of IgG subclass deficiency

Serum IgG contains 4 subclasses, IgG1 (60-66%), IgG2 (20-30%), IgG3 (less than or equal to 5%) and IgG4. Individual subclasses vary with respect to their physicochemical and biological properties. IgG subclass concentrations in serum are age dependent. IgG1 and IgG3 reach near to adult levels around the age of 3, IgG2 and IgG4 after the age of 6. Antibodies of certain specificities generally belong to a certain isotype (subclass) due to the isotype restriction. Patients with subclass deficiencies often suffer from recurrent infections. Those with IgG2 deficiency coften occurring with IgA and IgG4 deficiency) develop recurrent infection of the upper and lower respiratory tract often caused by pyogenic microorganisms (Haemophilus, Pneumo-(occus). Since early initiation of IVIG substitution therapy has a beneficial effect on long term prognosis the importance of early diagnosis is apparent.     

Role of immunoglobulin subclasses and specific antibody determinations in the evaluation of recurrent infection in children.

We studied humoral immune function in 267 children with recurrent respiratory infections referred to our immunology clinic to determine the most appropriate immunologic studies for evaluating recurrent infections in children. Of this highly selected population, 58% had a partial deficiency in one or more of the major immunoglobulin isotypes or IgG subclasses (defined as at least 2 SD below the normal age-adjusted mean). In none of the patients was there a total absence of an immunoglobulin isotype. The most common abnormality was partial IgA deficiency, which was found in one third of the patients. Twenty-six patients had only partial IgG subclass deficiencies, of which 20 were deficiencies of a single subclass. IgG1 was an isolated partial defect in three patients, IgG3 in five patients, and IgG2 and IgG4 were selective partial defects in six patients each. Tetanus toxoid and pneumopolysaccharide type 3 were the most immunogenic of the immunogens tested; hyporesponsiveness to pneumococcal polysaccharide types 7, 9, and 14 was common. Nineteen percent of the patients with normal immunoglobulin concentrations who were tested had lower-than-expected antibody titers; 42% of those tested with partial isotype deficiencies had deficient antibody responses. Of 25 patients with selective partial IgG subclass deficiencies or combined IgG subclass deficiencies, eight had antibody deficiencies. Our findings indicate that a high proportion of children referred to immunology clinics for recurrent infection have a demonstrable immunologic abnormality. Selective IgG subclass deficiency or a combined IgG subclass deficiency without an associated deficiency in a major immunoglobulin isotype is unusual. Identification of such patients is not predictive of the capacity to form antibodies to the antigens tested in this study and, in our opinion, adds little to the initial evaluation of immune function in such children.     

IgG subclasses in symptomatic IgA deficiency.

We evaluated IgG subclass levels in 11 symptomatic patients (ages between 3 and 22; mean 7.5 years) with IgA deficiency, seven with selective IgA deficiency, and four with low IgA levels. All patients had experienced three or more episodes of sinopulmonary infections a year. Combined IgG2-IgG4 deficiency was detected in two patients, IgG2 deficiency in one patient and IgG4 deficiency in two patients. Elevated IgG1 and IgG3 levels were detected in most of the IgG subclass deficient and sufficient patients. It is known that gammaglobulin replacement therapy reduces the frequency of infections significantly in IgG subclass deficiency. Although immunization against IgA is a risk in IgA deficiency, these patients can be treated with gammaglobulins containing low IgA.     

Undetectable IgG4 in immunoprecipitation: association with repeated infections in children?

A total of 210 patients with repeated infections were screened for IgG4 deficiency. In 30 patients (14%) IgG4 was undetectable by radial immunodiffusion (<30 mg/l). Of these patients 17 (57%) were less than 2 years of age. Concomitant IgA deficiency (IgA<0.05 g/l) was demonstrated in 11 cases (37%). IgG2 serum levels below the normal range were found in 26 children. IgG4 could be demonstrated at a concentration of 0.5–29 mg/l in all 30 patients using a more sensitive enzymelinked immunosorbent assay technique. Although a highly selected group of patients was investigated, the percentage of individuals without detectable IgG4 by immunodiffusion was in the same range as reported in the literature for healthy control persons. It is thus concluded that IgG4 serum reference levels have to be defined using more sensitive methods and that the observed severe infections are more likely to be connected with low serum IgG2 and/or IgA levels than undetectable IgG4 as measured by immunodiffusion.This work was supported by a grant from the Deutsche Forschungsgemeinschaft BA 872/1-1     

Disorders of immune function in children with selective IgA deficiency

Numerous additional alterations of immune function in patients with selective IgA deficiency (serum IgA less than 0.05 g/l) have been described. In this group of patients we have investigated the connection with allergic diseases and alterations of the other immunoglobulin isotypes. Sera of 44 children from 1 3/12 to 18 years were analysed. In all patients serum IgA was below the nephelometric detection limit of 0.05 g/l). Using a more sensitive ELISA, IgA could be detected in all sera in concentrations ranging from 10 micrograms/l to 0.04 g/l. 25 children (57%) revealed a profound elevation of IgG serum levels, in 27 (61%) IgM was elevated above the upper age related normal value. In 7 patients (16%) with normal IgG serum levels a combined IgG2-IgG4 deficiency was found. In most cases these patients had unusually frequent and severe infections. Total IgE serum levels were determined by a RIA technique. In addition, an IgE-mediated sensitization to the most common food and inhalation antigens was detected by a standardized procedure (Phadiatop, Pharmacia). In 9/44 children (20%) IgE was less than 2 U/ml (lowest detection limit), 27 patients (62%) revealed levels of 6-86 U/ml within the age-related normal range. In 8 patients (18%) total IgE was above 381 U/ml. Four patients demonstrated a sensitization to inhalants, specific IgE antibodies to nutritive antigens were detected in three children.(ABSTRACT TRUNCATED AT 250 WORDS)     

Serum immunoglobulin (IgG, IgM, IgA) and IgG subclass concentrations in healthy children: a study using nephelometric technique

Immunoglobulin (Ig) G, IgM, IgA and IgG subclass (IgG1, IgG2, IgG3, IgG4) concentrations were determined in more than 500 healthy Turkish children using nephelometric technique. These parameters were thought to be highly varied for different ethnic groups because of environmental and genetic factors. Methodology used in previous studies has been reported to affect age-related normal values. Serum IgG, IgM, IgA and IgG1, IgG2, IgG3, IgG4 levels were measured in 510, 491, 486, 542, 511, 515, and 545 healthy children, respectively. According to their age, the patients were divided into 14 groups. In contrast to most of the previous studies, age-related normal values for IgG4 levels were also obtained. In conclusion, it has been suggested that our study as an example for Caucasians using nephelometric technique will supply useful information about age-related normal serum immunoglobulin and IgG subclass concentrations.     

IgG2 deficiency in children with human immunodeficiency virus infection

Infection with the human immunodeficiency virus (HIV) induces a polyclonal B-cell activation. Despite elevated serum immunoglobulin levels, a significant deterioration of the antigen specific humoral immune response exists in most cases. We studied the influence of HIV infection on the serum levels of IgG subclasses in children. We investigated 76 children (aged 15 months to 18 years) with HIV-1-infection. Most children (88%) showed elevated serum immunoglobulin levels. IgA (87%) and IgM (74%) were more often above normal levels for age than IgG (60%). IgG subclass serum levels were significantly altered. The increase in total IgG was mainly due to a marked augmentation of the IgG1 fraction. In most cases IgG3 was simultaneously elevated. Ten children (13%) had very low IgG4 levels (<0.03 g/l). Out of 61 patients older than 2 years 8 (13%) had a profound IgG2 deficiency with normal or elevated total IgG. Four of them also had low IgG4 levels (<0.03 g/l). A correlation between IgG2 deficiency and HIV infection according to the Centres for Disease Control classification for acquired immunodeficiency syndrome could not be demonstrated (three patients with symptomatic and five with asymptomatic infection).     

Deficiency of IgG4 in children: association of isolated IgG4 deficiency with recurrent respiratory tract infection.

To study the relationship between serum IgG subclass deficiency and clinical host defense impairment, we reviewed the clinical and immunologic features of 123 patients with a history of recurrent infection who had been examined for immunodeficiency in our laboratory (group 1). We then compared immunoglobulin isotype levels with those in sera from 127 age-matched control subjects without recurrent infection from whom blood had been drawn for evaluation of atopy (group 2). There was a significantly higher prevalence of IgG4 deficiencies among patients with recurrent infections (17% vs 7%; p less than 0.02), solely because of a higher prevalence of isolated IgG4 deficiency (n = 9; 7.3%) than in atopic control subjects (n = 1; 0.8%; p less than 0.05); there was a comparable prevalence of multiple isotype deficiencies that included low levels of IgG4 (9.8% and 6.3%, respectively). All nine group 1 patients with isolated IgG4 deficiency had severe recurrent respiratory tract infections requiring multiple hospitalizations; in addition, five were atopic, five had asthma, and one had chronic diahrrea. Antibody responses to bacterial polysaccharide antigens were normal for age in all patients with isolated IgG4 deficiency; two had defective antibody responses to protein antigens. Isolated IgG4 deficiency appears to be associated with impaired respiratory tract defenses and may occur in the absence of an easily definable antibody deficiency state. This association suggests a physiologic defense role for mucosal IgG4.     

Transient IgG subclass deficiencies in newly diagnosed diabetic children

In 27 children (15 males and 12 females) with insulin-dependent diabetes mellitus (IDDM), aged 1.2–13.5 years (mean 9.9±3.6 years) we investigated immunoglobulins (IgG, IgA, IgM), IgG subclass levels and islet-cell antibodies (ICA) at diagnosis and at 6 and 12 months after disease onset. At diagnosis, IgG levels were lower than-2SD in 7 patients (26%), IgA in 1 (3.7%), IgM in 1 (3.7%). IgG subclass levels were below the 3rd percentile in 13 patients (48.1%); in particular IgG1 in 7 (26%), IgG2 in 3 (11.1%), IgG3 in 2 and IgG4 undetectable in 1 case. In 3 out of the 13 patients combined IgG1-IgG3, IgG1-IgG2 and IgG1-IgG4-IgA deficiencies were observed. ICA were >20 Juvenile Diabetes Foundation units in 17/27 patients. The HLA-DR2 frequency was higher in patients with IgG subclass deficiency than in patients with normal IgG subclass levels. During follow up, IgG levels normalized in 6 patients while IgA and IgM did not change. IgG1 normalized in 5 out of the 7 patients, IgG2 in all patients while IgG3 and IgG4 did not change. One year later ICA were still present in 8/27 patients. The hypogammaglobulinaemia and IgG subclass deficiencies observed in our patients could have either a genetic or an acquired basis.     

Selective IgA deficiency

Of 140 patients referred to the Pediatric Immunology Clinic during of 12-month period with the symptoms of recurrent infections or allergic respiratory illness, 21 (75%) were found to have selective IgA deficiency defined as serum concentration ≤ 5 mg. with normal levels of IgG and IgM. T lymphocyte number was reduced in the patients where as B cells with surface membrane IgA were increased. Autoantibodies and circulatory immune complexes were found more often in IgA-deficient subjects than in controls. Follow-up beyond the age of 9 years showed a spontaneous increase in serum IgA in 6, whereas 15 continued to have IgA deficiency. The latter group of children were characterized by more frequent infections, a higher prevalence of atopic disease, lower T cell count and serum IgG concentration, higher serum IgE level and a higher prevalence of food antibodies and immune complexes. These observations highlight the natural history and immunologic features of selective IgA deficiency.     

The outcome of patients with unclassified hypogammaglobulinemia in early childhood

Symptomatic hypogammaglobulinemia in childhood may be the initial finding of primary immunodeficiency (PID) or may be due to delay in maturation of immunoglobulin synthesis. The aim of this study was to review the clinical and laboratory records of patients with unclassified hypogammaglobulinemia and to evaluate whether these children experience changes in serum immunoglobulin concentrations during long‐term followup and have an exact diagnosis in natural course of disease. We reviewed the data of 412 patients who were diagnosed as PID with symptomatic hypogammaglobulinemia. Thirty‐seven patients with hypogammaglobulinemia [19 males (51.4%) and 18 females (48.6%), with a followup of 34.1 ± 22.0 months] who were not classified according to European Society for Immunodeficiencies diagnostic criteria were included in this study. The mean age at the beginning of the symptoms was 21.4 ± 20.6 months and the mean age at admission was 51.5 ± 25.8 months. The commonest clinical presentations were recurrent upper (94.6%) and/or lower (40.5%) respiratory infections, urinary infection (27%) and gastroenteritis (10.8%). Percentage of consanguinity was 8%. Of the initial 37 patients, 18 (48.6%) spontaneously corrected their immunoglobulin abnormalities during followup. Clinical symptoms of these patients were also improved. IgG, IgA and IgM levels reached to normal levels at ages 62.5 ± 21.8, 72.0 ± 11.2, 55.2 ± 7.8 months, respectively. In remaining 19 patients with undefined/unclassified hypogammaglobulinemia, three partial IgA deficiency, seven IgG subclass deficiency, two selective IgM deficiency and two common variable immunodeficiency (CVID) were diagnosed by long‐term monitoring of immunoglobulin levels. Five (13.5%) of the 37 unclassified patients could not be exactly diagnosed while two of them might have a T‐cell defect and three of them still had low IgG and IgA levels but adequate antibody responses against vaccine antigens. In conclusion, it is important to monitor symptomatic patients with hypogammaglobulinemia periodically. Some children may spontaneously correct their immunoglobulin abnormalities not in the first 30 months of age, but during the first decade of life and some of them may have a severe PID like CVID.     

IgG subclass deficiency in children with recurrent bronchitis

We studied the incidence of IgG subclass deficiency in children with recurrent bronchitis. Recurrent bronchitis was defined as three or more episodes a year during at least 2 consecutive years, of bronchopulmonary infection, productive cough with or without fever and/or diffuse rales by physical examination in the absence of asthma or atopy. Fifty three children were selected, of whom 30 (57%) were deficient in one of the IgG subclasses. None had an IgG1 deficiency. Nine (17%) were deficient in IgG2, 9 (17%) in IgG3 and 20 (38%) in IgG4. Isolated IgG subclass deficiencies were most frequently seen for IgG4 (14, 26%), less for IgG3 (6, 12%) and even less for IgG2 (4, 7%). Nine (17%) children were IgA deficient and 8 (15%) IgG deficient with a combined IgG subclass deficiency in 8 and 7 of them respectively. By subdivision into different age groups most patients were encountered in the youngest group. The mean content of IgG2, IgG3 and IgG4 in 3- to 4-year-old children with recurrent bronchitis was significantly lower than in the age matched controls. The mean value for IgG4 in the 5- to 6-year-olds was significantly lower than in the control group. This study demonstrates the correlation between recurrent bronchitis in childhood and IgG subclass deficiency. IgG subclass deficiency and recurrent bronchitis are both quite prominent phenomena in young children but rare in older children, suggesting a transient immaturity of the immune system as one of the possible pathogenetic factors. An IgA or an IgG deficiency is highly suggestive for the existence of a combined IgG subclass deficiency.     

Deficiency of humoral immunity in cartilage-hair hypoplasia

OBJECTIVE: Cartilage-hair hypoplasia (CHH), a metaphyseal chondrodysplasia, is usually associated with impaired cellular immunity. This study evaluates humoral immunity in patients with CHH. METHODS: The concentrations of immunoglobulins G, A, and M (IgG, IgA, and IgM) and IgG subclasses were studied in 20 patients. Data for 5 additional patients with recurrent infections were retrospectively reviewed. RESULTS: Seven of the prospectively evaluated patients (35%) had defective humoral immunity. Three patients had IgA deficiency. Four patients had IgG2 deficiency, accompanied by IgA deficiency, IgG4 deficiency, or both in 3 patients. IgG4 was low in most patients. Increased infections were usually associated with supranormal IgG and IgG1 and subnormal IgA, IgG2, or IgG4 concentrations. One retrospectively reviewed patient had severe hypogammaglobulinemia, and 3 had multiple IgG subclass deficiencies. CONCLUSIONS: Humoral immunity is impaired in CHH and contributes to the increased susceptibility to infections.     

Phenytoin-induced IgG2 and IgG4 deficiencies in a patient with epilepsy

A five-year-old girl with epilepsy and recurrent respiratory infections was investigated for serum IgG subclass concentrations. She was diagnosed as having a combined deficiency of IgG2 and IgG4 with a decreased serum concentration of IgA and IgG3 and was given replacement therapy with i.v. immunoglobulins. Since then, she has been free from respiratory infections. After phenytoin therapy was stopped, IgG subclass deficiency improved. This case describes the further action of phenytoin on the immune system, adding IgG subclass deficiency to the list.