Differences in clinical features found among countries
The differences in the clinical features found in the patients in the different regions analyzed are due more to the mode and method used for the clinical evaluation than to the patients themselves. For example:
The medical profile of the institution where the clinical analysis of patients is performed In Cuba, the pediatric referral hospital to which these children are referred is attached to a pediatric heart disease center; this may be the reason why in Cuba cases with deletion in 22q11.2 are showing higher rates of heart disease than in other studies. By contrast, in Russia, the selected patients come from a research center for mental illnesses, and cardiopathies are almost nonexistent, very mild and maybe not tested that intensely, but neurodevelopmental disorders are abundant. In one of the referral hospitals in Morocco, there is an infectious disease service. These diseases could cause fever and seizures. When apyretic seizures are detected, hypocalcemia/hypoparathyroidism studies are performed, if positive, then a direct referral is made to look for the deletion in 22q11.2.
The medical profile of the referring physician to the laboratory The recording of clinical features of a disease is different if we compare pediatricians, cardiologists, general clinicians, etc., with the subject matter expertise of a specialist in medical genetics. For example, in one of the laboratories in Morocco, all children from different parts of the country with heart disease (mainly CHD) are referred by non-geneticist physicians directly for molecular studies, and hence the high incidence of heart disease was detected. At the same time, there will also be many patients with cardiovascular malformations without the DGS-typical deletion tested. In India, there were no medical geneticists to make a clinical delineation of dysmorphic features and such prevailing clinical findings were only detected by ultrasound.
A correct algorithm for the study of these patients, designed by geneticists, seems to be preferable in DGS diagnostics. Hereby, priority is given to the clinical evaluation of the patient (seizures, suspected heart disease, facial dysmorphisms, neonatal sepsis) and complementary examinations (echocardiogram, thymus ultrasound, study of calcium levels, immunological studies) before a molecular test is performed. For example, in Cuba (where ecographic studies are prioritized because they are cheaper and more accessible to the population) as a way to alleviate the shortage of reagents for molecular studies of the deletion.
However, the great phenotypic heterogeneity present in the DGS syndrome makes a correct clinical delimitation difficult, which, together with the ethnic characteristics of each population, could potentially make the recognition of dysmorphic traits in affected individuals more problematic, as well [9,10,11,12,13].
In this study, a group of patients from different ethnic origins and geographic regions was gathered, identifying facial features of individuals with deletion del(22)(q11.2q11.2) in 68% of the patients despite the fact that several of them were at infant or prenatal stage. In order to make an early DGS syndrome identification, the clinical geneticist can rely on three fundamental pillars: endophenotype, facial findings, and functional disorders. The endophenotype is given by congenital defects present at birth including conotruncal cardiovascular, thymus hypoplasia, and palate defects. The facial findings most often can be associated besides with DGS, also with conotruncal anomaly face syndrome = Sedlackova syndrome (OMIM #217,095), autosomal dominant Opitz G/BBB syndrome (OMIM #300,000), and/or Cayler cardiofacial syndrome (OMIM %125,520) with a variety of signs. Considering facial features are variable and may not be present especially in persons of African/American heritage, still most important are ear dysplasia, nasal anomalies, asymmetric crying faces, micrognathia, a prominent nasal bridge, bulbous nose, and hypoplastic alae nasi. Functional disorders are associated with digestive system dysfunction (including palate), intellectual disability, behavior or mental disorders and also endocrine dysfunction of parathyroid and cellular CD3/CD4 immunologic presents in some cases even without thymes hypoplasia . In comparison—taken from the literature—in Brazil, at least 72.06% of the individuals present concomitantly two or more recurrent clinical manifestations indicative of deletion del(22)(q11.2q112.); most frequently seen in that study are velopharyngeal alterations and CHD .
Another predominant clinical feature of DGS is cardiovascular malformations (conotruncal or other) present in 68% of patients. Goldmuntz  states that cardiopathies are present in DGS individuals in a range between 60 and 80%. In recent studies the prevalence of heart disease has varied according to the age of diagnosis of this pathology; for example, Cancrini et al.  found a prevalence of CHD of 71%, if diagnosed before 2 years of age and only 22% after this age.
On the other hand, Poirsier et al.  found different prevalences of heart disease according to age a diagnosis was done first: they vary between 84% in fetal stage, 79% in neonates, 63% in children between 1 month and 2 years and 30% in children between 3 and 6 years.
The different diagnoses for CHD fall, according to Campbell et al. , who studied 1400 DGS patients, into three major groups: VSD (23%), tetralogy of Fallot (18%) and aortic arch anomalies (14%). This is in concordance with the present study for VSD (42%) and tetralogy of Fallot (22%), while aortic arch anomalies were almost absent here. Similar results were obtained by Poirsier et al.  doing a large multicenter study in France where VSD predominated in 30.7% and tetralogy of Fallot was found in only 17% of these patients. Nevertheless, Rozas et al.  found tetralogy of Fallot (20%) to be more frequent than VSD (14.7%). Goldmuntz  reviewed all these findings as (i) the prevalence of each subtype of heart disease is possibly associated with the age of the patients at the time of molecular diagnosis.
Combination of the most frequent clinical characteristics by age
In this study, in children under 2 years of age, cardiopathies were present in 67% of the individuals (30/45) together with facial dysmorphies, both being most useful from the clinical point of view to suspect this syndrome in children under 2 years of age. These two features could be complemented by determining possible hypocalcemia and/or hypoparathyroidism, as suggested by our findings. Accordingly, we suggest that the triad cardiopathy + dysmorphic features + hypocalcemia and/or hypoparathyroidism may be of great importance for the early diagnosis of DGS syndrome.
In patients older than 2 years, the most frequent clinical features were dysmorphic features + CHD + mental impairment + immunological disorders, which may be used for pointing the way toward DGS diagnoses. In general, young children with del(22)(q11.2q11.2) have delay in reaching motor milestones and emergence of language with a prevalence of nonverbal language at 2–3 years. Mental disabilities are frequently diagnosed after this age and include attention deficit, anxiety, perseveration, and difficulties in social interactions. Autism/autism spectrum disorders are reported in approximately 20% of DGS individuals. Behavioral abnormalities may set up before the age of 10 years, they are rarely observed, but may provide an opportunity for early intervention . Silva et al.  point out that psycho-pedagogical follow-up is a key aspect for recognition of individual school performance and psychosocial activities. Up to 40% of individuals with intellectual disability have a psychiatric disorder and these alterations manifest more frequently in adolescence, which make early diagnosis impossible if this may be the most evident characteristic of DGS in an individual.
In addition, a group of patients older than 2 years was identified, in which cardiovascular malformations are absent and distinctive features there were: dysmorphic features + immune disorders + mental dysfunction (10 cases). Goldmuntz  suggests that this may depend on the consistency with which a cardiac diagnosis is sought, and its classification is made. Especially in CHDs diagnosed after the first years of life, due to the absence of earlier relevant clinical features, those may not be evaluated that rigorously using advanced imaging techniques to detect clinically less significant cardiac features, such as aortic arch anomalies. Overall, the true CHD prevalence of DGS is difficult.
For prenatal diagnosis such an important identifier as ‘dysmorphic features’ cannot be evaluated; still fetal ultrasound findings are very useful for the diagnosis of this pathology. In this small series studied prenatally here, in seven fetuses anatomical malformations were found, predominantly as CHDs in 57% (4/7) and renal malformations in 29% (2/7). On the other hand, in one fetus, no malformation was detected by prenatal ultrasound, but it was known that the father was a carrier of the DGS deletion. In multicenter prenatal studies, the prevalence of the DGS was 1 in 100, when fetal anatomical anomalies were found, with a prevalence of heart disease. On the contrary, in those fetuses in which no anomalies were detected by prenatal ultrasound, DGS prevalence was 1 in approximately 1000 pregnancies [19, 20]. When it is known that one of the parents is a del(22)(q11.2q11.2) carrier, the risk for the offspring is 50%, and prenatal diagnosis should be recommended. For example, in this study, all four parents carrying the deletion had affected children.
However, it is interesting to note the point of view of Atli et al.  who consider phenotypic differences in individuals as a part of the syndrome and propose that also atypical anomalies should be taken into account to test for DGS, like developmental delay and intellectual disability along with CHD. Besides, deviating phenotypes in a family when identical deletions are present different causes should be considered as parental imprinting, unbalanced regulatory effects, polymorphisms not masked by recessive mutations or hemizygosity, environmental factors, and/ or stochastic events during morphogenesis.
In some studies also differences in DGS-critical region deletion size and affected regions are discussed [22, 23]. However, in the present study at least, for those patients studied by microarray, there was no evidence for variant deletions in 22q11.2. Still eight cases of this study with a phenotype that does not match the most frequent clinical feature groups can either be due to an ascertainment bias, that the clinical information provided may not be complete, or variant deletion sizes may have appeared (as those cases where only studied by FISH).
As in developed countries, testing methodologies are constantly improving, there is a greater likelihood of accurate diagnosis. Especially, it is increasingly common to replace descriptive, clinical diagnoses by direct molecular diagnosis . However, in developing countries, the traditional paradigm of genetic evaluation is maintained (the patient or tutor interrogation, the preparation of the genealogical tree, the clinical and dysmorphological examination, and the evaluation with complementary tests such as ultrasound and other laboratory tests) before performing costly genetic tests. Thus, genotype–phenotype correlations are more important for developing countries nowadays, than for industrial ones, even though saving costs in the health system is a topic in all societies.