American Journal of Kidney Diseases
Volume 52, Issue 6 , Page 1199, December 2008

In Reply to ‘The Pathogenesis of Acute Kidney Injury May Differ Between Elderly and Younger Patients’

  • Roland Schmitt, MD

      Affiliations

    • Hannover Medical School, Hannover, Germany
  • ,
  • Steve S. Coca, MD

      Affiliations

    • Yale University School of Medicine, New Haven, Connecticut
  • ,
  • Chiragh C. Parikh, MD, PhD

      Affiliations

    • Yale University School of Medicine, New Haven, Connecticut

Article Outline

 

We agree with Dr Schiffl1 that our results may have been confounded by a greater prevalence of chronic kidney disease or other comorbid conditions in the older population, as noted in our discussion. We also acknowledge that the phenotype of acute kidney injury (AKI) may be different in elderly compared with younger patients, and we agree that AKI should not be viewed as a reversible syndrome, but that chronic kidney disease caused by AKI is a predictor of long-term mortality.

Addressing the differences in data from Dr Schiffl and our own results, it is important to note that we had approximately 3,000 patients in our meta-analysis compared with only 226 surviving patients in the cohort of Dr Schiffl.2, 3, 4 Although the mean age of recruitment in the cohort of Dr Schiffl was 65 years, the survivors were younger (mean ages, 63 and 59 years). Thus, it appears that the death of very old patients may have added bias to the study. Dr Schiffl also states that age was not an independent predictor of nonrecovery in his analysis, but there are several reasons why this may be the case: lack of power with 100 events of partial recovery (outcome variable) and 9 independent variables along with several that have multiple subcategories, multicollinearity, homoscedasticity, or overfitting (information not provided for the latter 3). The multivariate analysis did not specifically examine the association of age with recovery, and age was one of the covariates in the model. Thus, the model-building techniques and other attributes of the multivariate model become important to consider before we can conclude that age is not related to nonrecovery. If the lack of association between age and recovery was caused by lack of statistical power, combining several studies in a meta-analysis approach can recover this effect.

Back to Article Outline

Acknowledgements 

Financial Disclosure: None.

Back to Article Outline

References 

  1. Schiffl H: The pathogenesis of acute kidney injury may differ between elderly and younger patients. Am J Kidney Dis 52:1198-1199, 2009
  2. Schmitt R, Coca S, Kanab M, et al. Recovery of kidney function after acute kidney injury in the elderly: A systematic review and meta-analysis. Am J Kidney Dis. 2008;52:262–271
  3. Schiffl H. Renal recovery from acute tubular necrosis requiring renal replacement therapy: A prospective study in critically ill patients. Nephrol Dial Transplant. 2006;21:1248–1252
  4. Schiffl H, Fischer R. Five-year outcomes of severe acute kidney injury requiring renal replacement therapy. Nephrol Dial Transplant. 2008;23:2235–2241

PII: S0272-6386(08)01427-3

doi:10.1053/j.ajkd.2008.09.017

Refers to article:

  • Recovery of Kidney Function After Acute Kidney Injury in the Elderly: A Systematic Review and Meta-analysis , 30 May 2008

    Roland Schmitt, Steven Coca, Mehmet Kanbay, Mary E. Tinetti, Lloyd G. Cantley, Chirag R. Parikh
    American Journal of Kidney Diseases August 2008 (Vol. 52, Issue 2, Pages 262-271)

  • The Pathogenesis of Acute Kidney Injury May Differ Between Elderly and Younger Patients

    Helmut Schiffl
    American Journal of Kidney Diseases December 2008 (Vol. 52, Issue 6, Pages 1198-1199)

American Journal of Kidney Diseases
Volume 52, Issue 6 , Page 1199, December 2008