Study: AKI alerts associated with lower mortality, LOS

Study: AKI alerts associated with lower mortality, LOS
share Share

A clinical decision support system (CDSS) that monitors blood creatinine levels in hospitalised patients to alert physicians to potential acute kidney injury (AKI) was associated with a small but significant reduction in mortality and hospital length of stay (LOS), according to research from the University of Pittsburgh and UPMC published in the Journal of the American Society of Nephrology.

The benefits of earlier detection of acute kidney injury include earlier intervention to mitigate loss of kidney function, and reduced hospital and long-term healthcare costs as a result of avoiding progression to severe and permanent kidney damage. Given the annual frequency of AKI in hospitalised U.S. patients of about 12% (2.2 million people), these results would translate into more than 17,000 lives and $1.2 billion saved per year, say the researchers.

The clinical decision support system is a computer program that reports in the Cerner EHR, used in 14 UPMC hospitals. If levels of serum creatinine rise too high or fast, an alert in the patient’s EHR informs physicians that AKI could be present. It also helps determine the stage of injury based on changes from the patient's baseline kidney function.


The researchers analysed the records of 528,108 patients between October 2012 and September 2015: 181,696 patients for 12 months before the alert system was implemented, and 346,412 patients for two years afterward. AKI was diagnosed in 64,512 patients (12.2%): 20,035 pre-implementation; 44,477 post-implementation

Crude mortality rate for patients with AKI:
      Pre-implementation: 10.2%
      Post-implementation: 9.4%
Crude mortality rate for patients without AKI: no change

Hospital LOS
 AKI patients
  •   Pre-implementation: 9.3 days
  •   Post-implementation: 9.0 days
Non-AKI patients: no change

The results also showed a decrease of 2.7 percent in dialysis rates. Even after adjusting for age and severity of illness, these changes remained highly significant. The researchers report a large effect on mortality in patients aged 60 years and older. They also note that the mechanisms by which hospital mortality, LOS and dialysis were reduced are unclear. They conclude that the results “support the development of CDSs to enhance early AKI detection but demonstrate that passive alerting will have only a limited effect on patient outcomes and more action-based CDSS will likely be needed to increase effect.”

Future developments

Senior author John Kellum, MD, professor of critical care medicine and director of the Center for Critical Care Nephrology at Pitt's School of Medicine told ICU Management & Practice in an email: "We are continuously refining the program based on performance. For example, the analytics behind determining baseline creatinine continue to evolve and we also have filters to reduce false positives. We also plan to add an active treatment module based on pharmacy evaluation for nephrotoxins.”

Source: UPMC

«« Better living quality for intensive care

Sleeping with the enemy: C.difficile infection in ICU »»


Al-Jaghbeer M, Dealmeida D, Bilderback A, Ambrosino R, Kellum JA (2017) Clinical decision support for in-hospital AKI. J Am Soc Nephrol, 29: doi. 10.1681/ASN.2017070765. 

Published on : Thu, 2 Nov 2017

Related Articles

It is estimated that 22–57% of critically ill patients develop acute kidney injury (AKI) during their hospitalisation. The increasing... Read more

Most cases of acute kidney injury (AKI) represent a manifestation of a system-wide disease, which should be addressed in all... Read more

Machine-Learning-Based Prediction Models for AKI

Acute Kidney Injury (AKI) is a common occurrence for critically ill patients in the ICU, and its early diagnosis has proven to... Read more

AKI, clinical decision support A clinical decision support system (CDSS) that monitors blood creatinine levels in hospitalised patients to alert physicians to potential acute kidney inj...

No comment

Please login to leave a comment...

Highlighted Products