Detecting Primary Immunodeficiency Disease in Children

Clinical Features That Identify Children With Primary Immunodeficiency Diseases

Subbarayan A, Colarusso G, Hughes SM, et al
Pediatrics 2011;127:810-816

Study Summary

The prevalence of primary immunodeficiency disease (PID) is approximately 1 in 10,000 children in the United States and 1 in 20,000 in Europe. It can be difficult to identify these children, and clinicians must have a high index of suspicion to recognize them at younger ages. In an informational campaign to educate both families and clinicians, patient and family support groups have promoted a group of 10 warning signs for PID. They include:

  • ≥ 4 new ear infections in 1 year;
  • ≥ 2 serious sinus infections in 1 year;
  • ≥ 2 months of oral antibiotic treatment with little effect;
  • ≥ 2 episodes of pneumonia in 1 year;
  • Failure of an infant to grow properly;
  • Recurrent deep skin or organ abscesses;
  • Persistent thrush in the mouth or other fungal infection on the skin;
  • The need for intravenous (IV) antibiotics to clear infections;
  • ≥ 2 deep-seated infections such as septicemia;
  • A family history of PID.

However, Subbarayan and coworkers comment that these warnings signs were not developed using population-based evidence to determine either frequency of these signs or their predictive ability. The researchers therefore examined the effectiveness of these warning signs by examining retrospective data from children evaluated for potential PID.

Data were obtained from referrals to 2 immunology clinics in northern England over a10-year period. Investigators reviewed records of more than 430 children who had a definitive diagnosis of PID. A comparison group included 133 children who presented to 1 of the centers with concerns of PID but who did not have PID after evaluation.

Of the 430 children with confirmed PID, the largest (242) group had T-lymphocyte dysfunction, with severe combined immunodeficiency being the largest subgroup. Complement deficiencies were evident in 22 children, 92 had B-lymphocyte dysfunction, and 74 had either neutrophil or monocyte dysfunction. The most frequent of the 10 warning signs in children who had a definable PID was a previous need for intravenous antibiotics in 56%. In children with a definable PID, 34% had a positive family history of PID compared with only 4% of children who did not have PID. In addition, 31% of the PID children exhibited failure-to-thrive compared with only 3% of children without definable PID. No differences in the number of deep-seated infections or episodes of pneumonia or abscesses were found between the children with and without definable PID. Frequencies of multiple otitis media episodes and sinus infections were lower in infants and children who had definable PID.

In regression analyses, 3 factors were predictive of a child having a definable PID. A positive family history of PID was associated with the highest relative risk (18; 95% confidence interval, 8-45). Requiring ≥ 2 months of oral antibiotics without improvement and failure to thrive were also significantly associated with a definable PID. A history of multiple episodes of acute otitis media or sinus infections was actually associated with a lower risk of definable PID in regression analyses. Family history remained a significant predictor of PID even among the PID subtypes. Failure-to-thrive was a significant predictor of both neutrophil and T-lymphocyte-related PID. Finally, of the 10 potential warning signs, family history was the only predictor of complement deficiency. The investigators conclude that a PID awareness initiative would be better targeted at hospital pediatricians and families with a history of PID rather than the general public. Their results offer a potential approach to refining the 10 warning signs to enable their use by clinicians to help identify PID in children.


Subbarayan and colleagues suggest using family history and a previous need for IV antibiotics as potential starting points for deciding whether to refer a child for PID evaluation. In Figure 2 of the article, they put forth a decision tree that first considers whether there is a family history of PID and then incorporates the need for IV antibiotics for bacterial sepsis. The presence of failure-to-thrive or congenital anomalies becomes a third ranked decision point. This decision tree has not been tested in a prospective fashion. Consequently, it should be viewed as hypothesis-generating and not as a validated decision approach to children with potential PID. Nonetheless, this is likely a very valuable step in developing a risk-based approach for future testing. The difficult issue for pediatric clinicians is that many normal children will have 1 or more of these 10 potential warning signs. Therefore, for the foreseeable future, identifying children with potential PID will require more of the art of medicine, a high index of suspicion, and perhaps a gut feeling for whether a child’s global presentation seems out of the ordinary compared with the other children encountered in practice. The investigators also offer a table that lists the screening investigations to consider for the 4 primary immunodeficiency types — an excellent table to keep handy for clinical reference.


Medscape Pediatrics © 2011 WebMD, LLC