To study the effect of ketoacidosis on measured 25-hydroxyvitamin D3 in children with new onset Type 1 diabetes.


Measurement of pH and bicarbonate levels was carried out in children with newly diagnosed Type 1 diabetes at presentation with ketoacidosis. 25-hydroxyvitamin D3 estimation was carried out at presentation (timepoint 1) and 1 month later (timepoint 2). There was no significant difference in the mean (±sd) 25-hydroxyvitamin D3 levels [35.39 (±25.79) vs 39.63 (±48.03) nmol/L; = 0.661) at the two timepoints in the study.


Correlation analysis revealed a positive correlation between bicarbonate levels and timepoint 1, i.e. the lower the bicarbonate levels, the lower were the timepoint 1 levels and vice versa (correlation coefficient 0.538, = 0.001). Timepoint 2 levels also showed a positive correlation with serum bicarbonate levels with a correlation coefficient of 0.379 (= 0.032). None of the variables other than bicarbonate,.(age, gender, BMI, pH or time), was found to have the predictive ability for timepoint1 levels. Similarly for predicting timepoint 2 levels, BMI was found to have independent predictive ability in addition to bicarbonate.


Severe ketoacidosis, as judged by bicarbonate but not pH, may transiently lower 25-hydroxyvitamin D3 levels in children with new onset Type 1 diabetes. Persistence of low 25-hydroxyvitamin D3 levels after resolution of ketoacidosis suggests a state of permanent vitamin D deficiency in our patient population.