Selective IgA deficit (SIgAD) belongs to the group of primary immunodeficiencies characterised by a deficit in the production of antibodies. Although SIgAD is the most frequent primary immunodeficiency (PID), its prevalence may vary from 1/163 to 1/18,500 cases, based on the studied population and the definition of SIgAD.1,2 These variations support a genetic predisposition, and new cases can appear sporadically or be transmitted in a dominant autosomal or recessive manner, or through polygenic inheritance. Thus, a 30% probability of SIgAD has been estimated if there is an affected relative,3 but no specific genetic defect has been identified, nor has a clear Mendelian inheritance pattern been proven, and expression and penetrance differences persist.2,4,5 Several genes potentially implicated in its pathogenesis have been identified, suggesting that SIgAD is probably a heterogeneous group of diseases and/or genetic anomalies similar to common variable immunodeficiency (CVID), notably transmembrane activator and calcium modulator (TACI), B-cell activating factor (BAFF-receptor), proliferation-inducing ligand (APRIL), cytotoxic T lymphocyte-associated protein-4-inducible co-stimulator (CTLA4-ICOS) and recombinant activating gene (RAG1), polymorphism, which could be associated a higher risk of presenting SIgAD, coeliac disease, and CVID.2,4,6–10 In addition to sharing genetic defects, familial aggregation and the known risk of SIgAD progressing to CVID suggest that they are probably part of a wide spectrum of the same disease.2,11 Furthermore, certain haplotypes HLA (A1, A29, B8, B14, DR3, DQ2) have been associated with a greater risk of SIgAD, although this association is not clear because these haplotypes are also associated with the autoimmune diseases frequently associated with this defect.2,12–14

Most patients (85%–90%) affected by SIgAD are asymptomatic. The clinical signs of symptomatic patients are wide ranging, including recurrent sinopulmonary infections, infections and gastrointestinal disorders, allergies, autoimmune processes and neoplasias, with more frequency than the general population.2,5,15,16

Although the diagnosis of SIgAD is usually made by analysis of serum immunoglobulins during the evaluation of recurrent respiratory infections, many other cases are diagnosed “accidentally”, as part of a laboratory evaluation for coeliac disease, allergy, or autoimmune disease.8

Prevalence of SIgAD is high among family members, and no studies have as yet been undertaken to compare the clinical and/or immunological difference among familial (FC) and sporadic (SC) cases. As a result, the severity, prognosis and need for complementary tests cannot be assessed. In this context, we set out to evaluate the prevalence of SIgAD in first-degree relatives (FDR) of SIgAD patients, determine whether screening family members is a useful strategy in the clinical management of this pathology, and compare the clinical and immunological characteristics in children diagnosed with sporadical SIgAD with those cases where there is an affected first-degree relative (FDR).

We found no evidence of the presence of more severe clinical phenotypes in familial cases of SIgAD in our series. We were only able to prove a clearly higher risk of admission due to respiratory pathology (most for asthma) in FCs, which could indicate a more severe phenotype; however, this was not confirmed when making an overall assessment of admissions for any cause in both groups.

In any event, a number of factors must be taken into consideration. Although patients with SIgAD are asymptomatic in most of the series reported in the literature,2,19 in our study all patients reported some kind of clinical sign, possibly due to selection bias in the centre where they received treatment. This may have contributed to the absence of significant clinical differences in our sample. On the other hand, loss of information is always a risk in retrospective studies, especially as regards minor clinical manifestations such as mild infections.

The distribution of infection by type did not differ to that reported in the literature5,16 and, despite the fact that overall infections were more frequent in SCs, there were no differences in the need for hospital admission, which implies a mild course in both groups. The low prevalence of allergic processes and intestinal, autoimmune and tumoral pathology in our sample compared with other series1,2,15,16,19–24 — probably related to the fact that it is a paediatric population — prevents us from reaching conclusions as regards the role of familial aggregation as a risk factor for developing this type of pathology. However, the high prevalence of positive ANA in patients in both groups with no evident autoimmune manifestations was striking and could be related to SIgAD as an epiphenomenon of autoimmune disorders that can increase the risk of developing an autoimmune disorder in adulthood, or as part of a compensatory response to infection and to a sustained condition, as shown by Fusaro et al.25 Likewise, total IgE levels were also higher in FCs despite not presenting a higher prevalence of clinically significant allergic processes.

Serum concentrations of IgG and IgM were significantly higher in FCs. There was no obvious reason for this, as prevalence of severity of infection or inflammation was not higher in this group.

Humoral response, measured through isohaemagglutinins and titres of ASLO, was poor in a significant number of patients (3 FCs and 6 SCs), so it would be convenient to make these determinations in all cases. The response to vaccines included in our vaccination schedule (tetanus, measles, mumps, rubella, H. influenzae and S. pneumoniae) was good in both groups, unlike the measles vaccine, where bad response was only reported in FCs. Irrespective of their group, we consider that these patients must be closely followed-up to assess whether these findings can predict the development of more clinically significant immunodeficiencies such as CVID, as reported in other publications.11

As regards the usefulness of family screening, it seems reasonable to affirm that patients detected based on the familial screening of paediatric patients with SIgAD present sufficiently significant clinical symptomatology to justify this strategy, particularly considering the high prevalence of familial cases shown in this and other studies.3,26,27

Thus, the most significant risk factor for the development of SIgAD shown in previous studies is the existence of a positive family history.28 Due to ethnicity-related genetic difference, the prevalence of affected family members no doubt differs in different regions of the world.3 In our analysis, the frequency of one or more FDR affected was 16%, diagnosed de novo in more than 70% of relatives. There was more than 1 case in the same family in 31% of cases. This percentage is similar to other series, which estimate figures of up to 32.4%, with an elevated prevalence of consanguinity not found in our sample3 and considerably higher than the general population, with a prevalence among Caucasians of 1 case in every 600.

Our study shows that the risk of SIgAD in FDRs of patients with this pathology was slightly higher in mothers (22%) than fathers (10%) and siblings (14.5%), although no major variations were found. These data are not consistent with previous studies, which report a higher risk in siblings.3,27,28 Vorechovský et al.28 analysed serum concentrations of Ig in close relatives of Swedish patients with SIgAD and CVID, estimating the relative risk for the siblings of patients with SIgAD to be approximately 50. However, parent of origin penetrance differences have been found in the transmission of SIgAD to children.29 Some investigations have found a higher prevalence of children with SIgAD born from mothers with this pathology than from fathers with the same defect.28,30 This type of inheritance with more predominance in maternal transmission could be related to mitochondrial defects and could explain in part the higher percentage of affected mothers found in our study, since the number of siblings (one brother/sister for each mother) is small in relation to other studies.3,27

Despite the fact that in our study only 14% of FDR-As were asymptomatic, this did not differ significantly from FDR-Hs, which is why it could be assumed that pathologies reported in FDR-As were similar to those in the general population. In 2008, Aghamohammadi et al.27 presented a study which affirmed that all relatives of patients with SIgAD who also had this defect were asymptomatic; however, they were not compared with the cohort of healthy relatives, which may explain our results. A recent case–control study comparing the symptoms of SIgAD patients with those of healthy patients found higher rates of infections, allergic phenomena and autoimmunity in patients with SIgAD26; we were unable to confirm this in our series. However, taking hospital admissions and need for chronic treatment as a marker of clinical severity would support the finding of Jorgensen et al.26

Although we were unable to show the significantly lower levels of IgG and IgM reported by other authors, who propose these variables as markers of risk for evolution to CVID, we detected 3 adult patients with hypogammaglobulinaemia who should be studied thoroughly.3,27 These findings, however, cannot be fully validated because IgG subclasses and antibody levels in family members were not evaluated.

In conclusion, we consider that the elevated prevalence of SIgAD in FDRs of patients with this pathology associated with a greater clinical severity, measured indirectly through admission rates and the need for chronic treatment, shows FDRs could benefit from screening programmes. Since few relatives will need to be screened to find one positive case, this could be a cost-effective prevention strategy. This would enable individuals with this type of immunodeficiency to start treatment or prophylaxis measures to prevent hospitalisation, or to simply attend regular check-ups due to the risk of evolution to CVID.

P. Soler-Palacín and E. Cobos-Carrascosa have contributed equally to the study.

The authors declare that there are no conflicts of interest.