On 2014 Oct 21, Emmanouil Giorgakis commented:

Interesting findings; however, the conclusions should be treated with caution. Drain Fluid Amylase (DFA) levels are function of both serum fluid amylase ( SFA) levels and the presence of a leak.. It would arguably be useful to introduce SFA levels into the equation and measure DFA/ SFA ratios rather than the absolute DFA levels: not only because a DFA>=100 would be clinically irrelevant if SFA levels were of the same range, but also because a DFA/SFA ratio is a figure independent of measuring units, thus applicable in various laboratory settings. Secondly, various authors have challenged the clinical usefulness of the biochemical diagnosis of an International Study Group on Pancreatic Fistula (ISGPF) Grade A Pancreatic Fistula ( PF), since such fistulae are transient and devoid of clinical sequealae or deviations in the clinical management. Furthermore, Cochrane meta-analysis published by Koti et. al. in 2010 demonstrated that the risk of PF is lower with the use of somatostatin analogues (SA), without this affecting the mortality rates. It would thus be crucial to clarify whether SA had been used perioperatively -and before a diagnosis of an established pancreatic leak was made-, on which case the outcome of the multivariate analysis and DFA1 cutoff should had been controlled for this variable among the rest. Thank you

On 2014 Oct 21, Emmanouil Giorgakis commented:

Interesting findings; however, the conclusions should be treated with caution. Drain Fluid Amylase (DFA) levels are function of both serum fluid amylase ( SFA) levels and the presence of a leak.. It would arguably be useful to introduce SFA levels into the equation and measure DFA/ SFA ratios rather than the absolute DFA levels: not only because a DFA>=100 would be clinically irrelevant if SFA levels were of the same range, but also because a DFA/SFA ratio is a figure independent of measuring units, thus applicable in various laboratory settings. Secondly, various authors have challenged the clinical usefulness of the biochemical diagnosis of an International Study Group on Pancreatic Fistula (ISGPF) Grade A Pancreatic Fistula ( PF), since such fistulae are transient and devoid of clinical sequealae or deviations in the clinical management. Furthermore, Cochrane meta-analysis published by Koti et. al. in 2010 demonstrated that the risk of PF is lower with the use of somatostatin analogues (SA), without this affecting the mortality rates. It would thus be crucial to clarify whether SA had been used perioperatively -and before a diagnosis of an established pancreatic leak was made-, on which case the outcome of the multivariate analysis and DFA1 cutoff should had been controlled for this variable among the rest. Thank you