We read with great interest the recent article by Balu, et al. [1]. However, in methodology, sample size considerations and calculations have not been mentioned.

As the sensitivity of gold standard has been mentioned as 88%, we calculated the sample size for 80% with a deviation of 5% on either side and the calculations show that sample size in this study is woefully inadequate [2]. Thus the study was inadequately powered to assess the validity of the proposed screening tests. As PPV depends upon prevalence, likelihood ratio is a better measure to overcome this inadequacy which has not been calculated in this study. Authors also state that more training (how much) would be required to get a better result, which raises the question of internal validity. In the diagram describing association of pulse oximetry with clinical evaluation and echocardiography it seems that all newborns went through pulse oximetry first followed by clinical examination but authors initially mentioned the reverse sequence. The appropriate sequence of clinical examination followed by oximetry has also been described earlier [3].