Dear Editor, I read with great interest the study by Chandrashekhar and Jeet1 comparing stroke volume variation (SVV) and plethysmography variability index (PVI) for goal-directed fluid therapy. Chandrashekhar and Jeet1 have used either the SVV or PVI to guide fluid administration in each group of patients. They have also plotted the receiver operating characteristics (ROC) curve for each of these indices. The ROC curve evaluates the discriminatory ability of a diagnostic test to correctly pick diseased and non- diseased subjects. 2 It plots the sensitivity (true positive/true positive + false negative) on the y-axis and 1-specificity 1- (true negative/true negative + false positive) on the x-axis. Here, the diagnostic test is SVV or PVI and the outcome measure is response to a fluid bolus (in the form of reduction in SVV or PVI). However, Chandrashekhar and Jeet1 state in their methodology that they administered a fluid bolus when the index (SVV or PVI) was >11%. They have not used any other end point for administration of a fluid bolus. Thus, it is not possible to identify the false negatives and true negatives. That is, the patients who were fluid responsive despite a low value of SVV or PVI (false negatives) cannot be identified. Therefore, the ROC curve cannot be applied to this study. The fluid requirement in major abdominal oncosurgeries also depends on variables like pre-operative bowel preparation, pre-operative hemoglobin level, and intraoperative blood loss. Chandrashekhar and Jeet1 have not presented the data comparing these variables among the two groups which could confound the results. Chandrashekhar and Jeet1 state that they maintained a train of four (TOF) ratios of 50% during the surgery which may be a typing mistake. In conclusion, I congratulate Chandrashekhar and Jeet1 for using goal directed fluid therapy although their statistical analysis is questionable. Financial support and sponsorship Nil. Conflicts of interest There are no conflicts of interest.
Vidhya Narayanan (Wed,) studied this question.