Jump to comment:

• Reply to e-letter by Xue et al - Use of preoperative neutrophil gelatinase-associated lipocalin to predicts acute kidney injury and mortality after cardiac surgery
  Derek Hausenloy, Heeraj Bulluck
  Published on: 29 January 2018
• Use of preoperative neutrophil gelatinase-associated lipocalin to predicts acute kidney injury and mortality after cardiac surgery
  Fu-Shan Xue, Chao Wen and Gui-Zhen Yang
  Published on: 29 January 2018

• Published on: 29 January 2018

  Reply to e-letter by Xue et al - Use of preoperative neutrophil gelatinase-associated lipocalin to predicts acute kidney injury and mortality after cardiac surgery

  • Derek Hausenloy, Cardiologist Duke-National University Singapore
  • Other Contributors:
    • Heeraj Bulluck, Cardiologist

  We thank Xue et al for their interest in our article (1). The KDIGO classification (2) comprises 3 criteria in the diagnosis of acute kidney injury (AKI), namely an increase in serum creatinine (SCr) by ≥0.3 mg/dl (≥26.5 μmol/l) within 48 hours; or an increase in SCr to ≥1.5 times baseline, which is known or presumed to have occurred within the prior 7 days; or urine volume <0.5 ml/kg/h for 6 hours. In our study, we had collected serial blood samples for the first 72 hours and the second KDIGO criterion (an increase in SCr to ≥1.5 times baseline) was applied over the 72 hours period to assess for AKI. As for the impact of perioperative fluid balance, this is not currently part of the recommendation, and the available evidence quoted (3) was taken from a retrospective, single-center study using the AKIN classification for AKI. Unfortunately the perioperative fluid balance was not collected in our study and we could not take this into account.
  We did look at the Receiver Operating Characteristic curve analysis using sNGAL as a continuous variable. The area under the curve (AUC) was 0.57 (95%CI 0.54-0.60), very similar to that of sNGAL tertiles reported in our paper.(1) The diagnostic performance of sNGAL alone was quite low and therefore we did not provide cut-off values/sensitivity/specificity and predictive values. Expressing sNGAL as quartiles is more practical from a clinical point of view and we showed that by adding clinical factors, the c-statistic improved...

  Show More

  We thank Xue et al for their interest in our article (1). The KDIGO classification (2) comprises 3 criteria in the diagnosis of acute kidney injury (AKI), namely an increase in serum creatinine (SCr) by ≥0.3 mg/dl (≥26.5 μmol/l) within 48 hours; or an increase in SCr to ≥1.5 times baseline, which is known or presumed to have occurred within the prior 7 days; or urine volume <0.5 ml/kg/h for 6 hours. In our study, we had collected serial blood samples for the first 72 hours and the second KDIGO criterion (an increase in SCr to ≥1.5 times baseline) was applied over the 72 hours period to assess for AKI. As for the impact of perioperative fluid balance, this is not currently part of the recommendation, and the available evidence quoted (3) was taken from a retrospective, single-center study using the AKIN classification for AKI. Unfortunately the perioperative fluid balance was not collected in our study and we could not take this into account.
  We did look at the Receiver Operating Characteristic curve analysis using sNGAL as a continuous variable. The area under the curve (AUC) was 0.57 (95%CI 0.54-0.60), very similar to that of sNGAL tertiles reported in our paper.(1) The diagnostic performance of sNGAL alone was quite low and therefore we did not provide cut-off values/sensitivity/specificity and predictive values. Expressing sNGAL as quartiles is more practical from a clinical point of view and we showed that by adding clinical factors, the c-statistic improved to 0.69. We are aware that the value of 0.69 is not ideal for clinical application and our work was a post-hoc analysis of a randomized controlled trial (4) consisting of high-risk patients (EuroSCORE≥5). Therefore, further work remains to be done in a larger number of all-comers going for cardiac surgery to improve the identification of patients at risk of AKI.

  References
  1. Bulluck H, Maiti R, Chakraborty B, et al. Neutrophil gelatinase-associated lipocalin prior to cardiac surgery predicts acute kidney injury and mortality. Heart 2017 doi: 10.1136/heartjnl-2017-311760
  2. James M, Bouchard J, Ho J, et al. Canadian Society of Nephrology commentary on the 2012 KDIGO clinical practice guideline for acute kidney injury. American journal of kidney diseases : the official journal of the National Kidney Foundation 2013;61(5):673-85. doi: 10.1053/j.ajkd.2013.02.350
  3. Moore E, Tobin A, Reid D, et al. The Impact of Fluid Balance on the Detection, Classification and Outcome of Acute Kidney Injury After Cardiac Surgery. Journal of cardiothoracic and vascular anesthesia 2015;29(5):1229-35. doi: 10.1053/j.jvca.2015.02.004
  4. Hausenloy DJ, Candilio L, Yellon DM. Remote Ischemic Preconditioning and Cardiac Surgery. The New England journal of medicine 2016;374(5):491-2.

  Show Less

  Conflict of Interest:
  None declared.

  • Back to top
• Published on: 29 January 2018

  Use of preoperative neutrophil gelatinase-associated lipocalin to predicts acute kidney injury and mortality after cardiac surgery

  • Fu-Shan Xue, Anesthesiologist Plastic Surgery Hospital, Chinese Academy of Medical Sciences and Peking Union Medical College, Beijing, China
  • Other Contributors:
    • Chao Wen, Anesthesiologist
    • Gui-Zhen Yang, Anesthesiologist

  We read with great interest the recent article by Bulluck and colleagues regarding use of preoperative serum neutrophil gelatinase-associated lipocalin (sNGAL) to predict acute kidney injury (AKI) during hospitalisation and 1-year cardiovascular and all-cause mortality following adult cardiac surgery. They showed that preoperative sNGAL was an independent predictor of postoperative AKI and 1-year mortality. Although the valuable study has been actualized, two issues in methodology seem important to avoid any optimistic interpretation or misinterpretation of results.
  First, when using the KDIGO criteria to define and grade AKI, Bulluck et al1 used a time window of 72 h to include patients with different serum creatinine (sCr) increases from baseline, rather than 48 h, as specified by the guideline. Furthermore, it was unclear whether the sCr levels used for diagnosis and staging of AKI had been corrected based on perioperative fluid balance. The available evidence shows that not adjusting sCr levels for fluid balance may underestimate incidence of AKI after cardiac surgery, as a positive perioperative fluid balance may dilute sCr.2
  Second, this study only assessed the associations of preoperative sNGAL levels with the risks of postoperative AKI and 1-year mortality, but did not provide the true predictive performances of preoperative sNGAL. To determine discriminative ability of preoperative sNGAL for adverse postoperative outcomes, the receiver operating charac...

  Show More

  We read with great interest the recent article by Bulluck and colleagues regarding use of preoperative serum neutrophil gelatinase-associated lipocalin (sNGAL) to predict acute kidney injury (AKI) during hospitalisation and 1-year cardiovascular and all-cause mortality following adult cardiac surgery. They showed that preoperative sNGAL was an independent predictor of postoperative AKI and 1-year mortality. Although the valuable study has been actualized, two issues in methodology seem important to avoid any optimistic interpretation or misinterpretation of results.
  First, when using the KDIGO criteria to define and grade AKI, Bulluck et al1 used a time window of 72 h to include patients with different serum creatinine (sCr) increases from baseline, rather than 48 h, as specified by the guideline. Furthermore, it was unclear whether the sCr levels used for diagnosis and staging of AKI had been corrected based on perioperative fluid balance. The available evidence shows that not adjusting sCr levels for fluid balance may underestimate incidence of AKI after cardiac surgery, as a positive perioperative fluid balance may dilute sCr.2
  Second, this study only assessed the associations of preoperative sNGAL levels with the risks of postoperative AKI and 1-year mortality, but did not provide the true predictive performances of preoperative sNGAL. To determine discriminative ability of preoperative sNGAL for adverse postoperative outcomes, the receiver operating characteristic curve analysis should be performed to provide the optimal cutoff values of preoperative sNGAL for postoperative AKI and mortality as well as its sensitivity, specificity, and positive and negative predictive values. The optimal cutoff value is the one that has the highest sensitivity and specificity combined. By providing the predicted probabilities and observed frequencies for postoperative AKI and mortality based on the cutoff values of preoperative sNGAL, the readers can estimate whether there is a good overall agreement between predicted probabilities and observed frequencies in the development and the validation sets. This study identified preoperative sNGAL as a predictor of AKI, but c-statistic was only improved to 0.69 when controlling perioperative risk factors affecting postoperative myocardial and kidney injuries. Traditionally, a c-statistic of 0.5 means random occurrence, whereas a c-statistic > 0.7 signifies adequate discrimination and > 0.8 indicates strong discrimination.

  REFERENCES
  1 Bulluck H, Maiti R, Chakraborty B, et al. Neutrophil gelatinase-associated lipocalin prior to cardiac surgery predicts acute kidney injury and mortality. Heart 2017 Aug 9. DOI: 10.1136/heartjnl-2017-311760.
  2 Moore E, Tobin A, Reid D, et al. The impact of fluid balance on the detection, classification and outcome of acute kidney injury after cardiac surgery. J Cardiothorac Vasc Anesth 2015; 29:1229-35.
  3 Merkow RP, Hall BL, Cohen ME, et al. Relevance of the c-statistic when evaluating risk-adjustment models in surgery. J Am Coll Surg 2012; 214:822-30.

  Show Less

  Conflict of Interest:
  None declared.

  • Back to top