Cystatin C is a sensitive marker of prognosis acute heart failure

Susan Jeffrey

September 06, 2005

Sep 6, 2005

Stockholm, Sweden - A new study suggests that cystatin C, a novel marker of renal function, may be a more sensitive marker of prognosis in patients with acute heart failure than established markers such as creatinine, BNP, NT-proBNP, or troponin T.

Dr Johan PE Lassus

In this analysis, the highest six-month mortality was seen in the highest tertile for cystatin C and was significantly higher than mortality in the lowest tertile, the authors report. Trends toward higher mortality with higher levels were seen with the other markers as well, but none reached statistical significance.

"Cystatin C seems to be a promising new risk marker in patients admitted with acute heart failure," lead author Dr Johan PE Lassus (Helsinki University Central Hospital, Finland) concluded.

Their results were presented here at the European Society of Cardiology Congress 2005.

A novel marker of risk

Cystatin C is a novel marker of renal function that has generated some interest for its potential utility in predicting risk in a variety of populations. Earlier this year in a paper in the New England Journal of Medicine, Shlipak and colleagues compared creatinine and cystatin-C levels as predictors of mortality in 4637 participants in the Cardiovascular Health Study, a cohort of elderly people living in the community [ 1 ].  They found that higher cystatin-C levels were directly associated, in a dose-dependent manner, with a higher risk of death from all causes, while serum creatinine showed a J-shaped association, with only a small group of about 10% of the population falling into the high-risk category.

In this report, Lassus and colleagues studied a population of 622 patients admitted for acute heart failure in Finland, who were part of the multicenter FINN-AKVA study. Patients were elderly, with a mean age of 75 years. The gender distribution was equal, and 50% of the patients had a history of heart failure, so half were newly diagnosed on admission. One third of patients had a well-preserved ejection fraction and could be classified as diastolic heart failure, Lassus said.

They analyzed a variety of biomarkers, dividing patients into tertiles from lowest (tertile 1) to highest (tertile 3) levels. Markers included BNP, NT-proBNP, troponin T, and markers of renal function including serum creatinine, from which they calculated the glomerular filtration rate (GFR) using the Cockcroft-Gault formula.
"What we found in mortality analysis is that cystatin C was the strongest predictor of outcome at six months, and the tertiles of cystatin C divided the patients into low-, medium-, and high-risk groups," Lassus told renal wire .

6-month mortality by tertiles of cystatin C

6-month mortality (%)
Tertile 1
Tertile 2
Tertile 3

Comparing the third with the first tertile for each of the markers, only the comparison using cystatin C reached statistical significance (p=0.006), although trends were seen with both serum creatinine and eGFR. Lassus speculated during the session that this may reflect the relatively small sample size, although it also implies that cystatin C may be a more sensitive marker than these other, more standard measures.
"This analysis I presented today was done on a subgroup of the whole population who had laboratory analysis at hand when we did this," he said. "Now we have three times as a large population, the analysis has been done, and it is quite similar, so there is a difference in these tertiles analyzed by cystatin C. We see that the sensitivity of cystatin C to predict outcome is better than for creatinine or the natriuretic peptides.

"At the moment we need to focus on patients with renal failure or renal dysfunction, and I think that cystatin C might be the more sensitive marker of renal dysfunction," Lassus told renal wire . "It has been stated that it could have an independent prognostic impact beyond that of renal function, but I think that's really quite a new finding, and we have to investigate more before we can make any changes in our clinical practice based on that."

Underestimating GFR

In a separate report here during the same session, researchers in the departments of cardiology and nephrology at the University Medical Center Groningen, the Netherlands, found that the severity of heart failure can substantially affect estimated GFRs based on serum-creatinine-derived formulas.

Dr Tom DJ Smilde presented the data in 80 patients with congestive heart failure, showing that compared with the gold-standard GFR measure using 125l-iothalamate clearance, or the "true" GFR, GFRs calculated using the Cockcroft-Gault, the Modification of Diet in Renal Disease (MDRD) formula, as well as the simplified MDRD were all significantly different from the gold-standard method.

Mean GFR by various methods in patients with chronic heart failure

GFR measurement
Mean GFR (mL/min per 1.73 m 2 )
GFR (gold-standard method)
75 +27
Cockcroft-Gault method
67 +21
62 +19
Simplified MDRD
63 +18

To download tables as slides, click here

All of the formulas underestimated GFR in patients in New York Heart Association (NYHA) classes 1-3, mild to moderate heart failure, but the simplified MDRD and Cockcroft-Gault calculations actually overestimated GFR in class 4 patients with severe heart failure.
This last finding can be dangerous, Smilde told doctors here, since patients with severe heart failure may require renal-replacement therapy but will not receive it because their GFR has been overestimated.

"I think it's very important to recognize that creatinine and formulas that calculate GFR based on creatinine do have several limitations, and that was very well pointed out by Dr Smilde," Lassus said of these findings. "So that reemphasizes the need for a better marker of renal function that we can use in clinical practice. We cannot, due to patient characteristics and costs, use the gold standard to measure it with radionuclide agents."


  1. Shlipak MG, Sarnak MJ, Katz R, et al. Cystatin C and the risk of death and cardiovascular events among elderly persons. N Engl J Med 2005; 352:2049-2060. Abtract



Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: