Ketoacidosis as a debut to type 1B diabetes mellitus in a patient with Turner's syndrome

In Turner syndrome (TS), mortality is three times higher as compared to the population with a normal karyotype. This is partly explained by the increasing incidence of diabetes mellitus (DM) and cardiovascular diseases.¹ There is an increased risk of developing both type 2 DM (relative risk, 4.4) and type 1 DM (DM1) (relative risk, 11.6) as compared to the general population.² Development of frank DM during childhood is however exceptional, and is usually more commonly associated to treatment with GH or sex hormones.^2,3

The unusual case of a two-year-old patient with DM1 of a non-autoimmune origin in whom TS was diagnosed is reported. Few patients with TS in whom DM1 has developed in childhood have been reported,^3–6 and this is the youngest known.

Our patient was a Roumanian girl aged 2 years and 4 months with polyuria, polydipsia, and weight stagnation for the past two weeks. She weighed 8.100 kg (−3.6 SDS) and measured 77.5 cm in height (−3.8 SDS). The patient had a unique phenotype, with anteverted, low-set pinnae, slight hypertelorism, and mandibular hypoplasia. Hyperglycemia (682 mg/dL) and metabolic acidosis (pH 7.13, pCO₂ 20 mmHg, HCO₃ 6.7 mmol/L, BE −20.6 mmol/L) were detected, which allowed for diagnosing diabetic ketoacidosis. Diabetes work-up was completed by measurement of HbA1c (11.1%) and C peptide (0.25 ng/mL; normal: 0.90–4 ng/mL), the latter by immunoluminescence. Tests for diabetes antibodies were requested (indirect immunofluorescence method: islet cell antibodies; radioisotopic method: insulin, glutamic acid decarboxylase, and tyrosine phosphatase antibodies) and they were all negative. There was no family history of autoimmune diseases. Type B DM1 was diagnosed based on the lack of evidence of autoimmunity.

Because of the unique features and low height, karyotype was requested and was found to be consistent with TS. A cytogenetic study of 15 metaphases was performed from a culture stimulated with phytohemagglutinin, with a resolution of 400 bands, and an isodicentric X chromosome with long arms was seen, which was confirmed by the band technique C: 46,X,psu idic(X)(p11.2)[15].

TS is associated to autoimmune conditions. Thus, when DM1 is diagnosed it appears logical to think that it is of the 1A type; however, some authors state that autoantibodies are more frequently negative in TS. This statement may however be conditioned by classification of DM based on the presence or absence of insulin dependence, which may lead to a wrong diagnosis of DM1 in patients with type 2 DM.^3,7 On the other hand, some authors postulate the existence of mechanisms of destruction of pancreatic beta cells (PBCs) other than autoimmunity.²

Insulin deficiency secondary to PBC dysfunction and worsening with age has been reported in TS. At the earliest ages, this insulin deficiency is compensated by increased insulin sensitivity, but this would gradually decrease over time, which would lead to carbohydrate intolerance or even DM. Pathogenesis of this condition is independent from increases in body mass index and hypogonadism associated to TS, and is similar to pathogenesis of the mature-onset diabetes of the young (MODY), secondary to haploinsufficiency of some transcription factors affecting PBC function.⁴

A potential relationship has been seen between metabolic diseases and presence of an isochromosome of the long arm of the X chromosome (iXq).⁴ Schoemaker et al.¹ investigated causes of death in TS and found that in iXq karyotypes increased mortality was related to DM. It is hypothesized that there are genes in X chromosome involved in PBC function, and that haploinsufficiency of such genes could account for DM occurrence. This is also supported by the close similarity between type MODY diabetes and changes in carbohydrate metabolism in patients with TS.⁸ There are also studies reporting a greater prevalence of DM1 related to findings made in the short arm of the X chromosome (Xp). Bakalov et al.⁹ suggested that the greater incidence of DM in TS may be due to haploinsufficiency of genes located in chromosome Xp, so that DM incidence will be higher in patients with Xp monosomy (45X; delXp; iXq), with a normal incidence in long arm deletions (delXq), where both Xp arms are preserved. However, the greater incidence of DM in iXq patients may also be explained by overexpression of genes located in chromosome Xq which would escape inactivation, and which would be related to PBC function, and to another series of systemic processes which would promote a pro-inflammatory state. These statements are supported by a study conducted in 2008 showing a higher mortality rate in iXq patients,¹ and by the fact that there is also a greater incidence of DM in patients with Klinefelter syndrome (47XXY) and in 48 XXYY patients, in whom there are supernumerary copies of Xq.¹⁰

Association of TS and DM1 may actually be due to mechanisms inherent to chromosomal findings. Association of TS to autoimmunity is however widely known, and a potential, currently undetectable autoimmune basis cannot therefore be ruled out with full certainty. Regular autoantibody monitoring is therefore recommended.

Association of DM to genetic syndromes provides unique information for studying the genetic cases of carbohydrate metabolism. There is still a long road ahead before knowing the pathophysiological and genetic bases justifying DM1 occurrence in TS. New advances in genetic testing, as well as creation of multicenter databases, will allow for more complete understanding of this syndrome and PBC pathophysiology.

Conflict of interest

The authors state that they have no conflicts of interest.