Slide 1

Thank you. Dear moderators, dear colleagues, the discovery that inflammation is related to a very high risk for death and cardiovascular complications is perhaps one of the most important observations made during the last decade in ESRD research. The link between the prototypical CRP, the inflammatory protein for excellence and poor outcomes is exemplified in this paper which is the joint effort of 3 European groups with major interest in this issue.

Slide 2

In this paper that has been shown many times during this congress the dialysis patients in the first quartile that is with the lowest value of CRP had a 5-year survival that was 80%. While patients in the fourth quartile, those with the highest value of CRP, had a survival of about 40%.

Slide 3

One of the reasons why inflammation is related to poor outcomes is that cytokines are pro-catabolic; they induce loss of appetite, loss of lean and fat body mass. This pro-catabolic effect of cytokines is a quite strong  pathophysiological perturbation and  it triggers  adaptive responses aimed at restoring energy balance.
Studies of the adaptive response to inflammation have been extensively done in patients with sepsis in the intensive care unit and in these patients the function of the thyroid gland was suppressed and mainly was suppressed free-T3. This condition has been named euthyroid sick syndrome because it was thought that this is an innocent condition, a sort of answer to the septic state. This euthyroid sick syndrome has been described also in the inflamed patients with heart failure or liver failure. Moreover, 15-30% of ERSD patients present with euthyroid sick syndrome. The problem is particularly relevant because there is emerging evidence that euthyroid sick syndrome may not be an innocent adaptation as originally thought.

Slide 4

Evidence that low levels of the active thyroid hormone, FT3, may be detrimental in the setting of inflammation  emerged in a study in patients  with heart failure. In this study patients with heart failure were stratified  according to FT3. Remarkably, 1 year survival was almost 100% in those with normal free T3  but 80%, a 20% mortality rate, in those  with low FT3. Thus low FT3 in the setting of inflammation is known to herald poor outcomes.

Slide 5

This phenomenon is highly relevant in patients with heart disease. Indeed, unlike in other organs, the metabolism of myocardial cells uniquely depends on circulating T3 because these cells are unable to transform circulating T4 into the active hormone, T3. This is nicely demonstrated in this experiment  in “in vitro”  perfused rabbit hearts. As you can see  LV function recovery after ischemia  was much enhanced if T3 was added to the perfusate as compared to control experiment.

Slide 6

Today I’m going to speak about the steady state relationships between inflammatory cytokines and free-T3 in dialysis patients. Then I will look at dynamic relationships between inflammatory cytokines during acute episodes of inflammation and free-T3. Then we are going to look at the relationships between free-T3 and the left ventricular mass and function in dialysis patients. Last but not least we will look at the free-T3 survival in ESRD patients.

Slide 7

With this background in mind we studied 200 hemodialysis patients, 31 healthy subjects and 262 euthyroid subjects from the general population. This survey was published in  JASN almost 2 years ago and, as you can see,  circulating levels of plasma fT3 in dialysis patients were much lower than in age and sex matched controls and in clinically euthyroid patients with normal renal function.

Slide 8

Because inflammation influences thyroid function, low plasma fT3 in ESRD may be an expression of the inflammatory state of these patients. Therefore we investigated the relationships between  IL6 and CRP which are two solidly established markers of the malnutrition- inflammation syndrome, and the most metabolically active hormone ( free T3) in hemodialysis patients. As you can see, IL6 and CRP were inversely related to fT3 plasma levels: the higher the IL6 and CRP, the more severe the inflammation, and the lower the fT3. Inflammation markers apart, serum albumin is another, inverse, acute-phase reactant. Of note is that fT3 was strongly and directly associated to serum albumin: high  serum albumin high plasma fT3.

Slide 9

A complementary analysis correlating fT3 and previous CV events in dialysis patients revealed a graded inverse relationship between the number of past CV events and the plasma concentration of fT3. This strong gradient type relationship between CV events and fT3, could mean that depressed fT3 may be an unsuspected player in the high risk associated with inflammation in these patients, an issue that deserves to be explored more extensively in specifically designed studies.

Slide 10

Now, the second point. The dynamic relationships during acute episodes.

Slide 11

These studies were performed in 17 consecutive CKD patients  admitted to the hospital with high fever. The causes of inflammation/ infection were the most disparate, from bronchitis to infected central catheters. IL6  CRP as markers of inflammation, and the active thyroid hormone (fT3) were measured at the outset of the inflammatory process and repeated when the process was either attenuated or resolved itself. As you can see IL6 and CRP were elevated in the zenith of the illness and both of them decreased substantially at the nadir that is when the inflammation/infection was partially or completely resolved. By contrast fT3 was depressed during the zenith and increased substantially after the resolution of the acute episode process. The fact that we found that fT3 was reduced to a great extent during intercurrent inflammatory processes and that this alteration resolves itself as inflammation fades away, documents that inflammation acutely interferes with thyroid function in patients with CKD and that this interference is fully reversible independently from the GFR.

Slide 12

The third point: the relationships between free-T3 and left ventricular mass and function in haemodialysis patients.

Slide 13

As previously said, the heart is highly dependent on circulating T3  because the myocardiocyte has no deiodinase activity to make up T3 locally. Furthermore Ft3 is involved in the control of myocardial contractility and vascular tone and low plasma fT3 is frequently observed in patients with heart diseases. ESRD is a high risk condition and heart disease is a very common complication in patients on dialysis. In our dialysis cohort the prevalence of left ventricular hypertrophy and LV dysfunction, the strongest predictors of death in these patients, was 77% and  47% respectively.

Slide 14

To look into this problem we investigated the relationship between fT3  and Midwall Fractional Shortening, a measure of cardiac contractility, and Left Ventricular Mass. As you can observe, MWFS was significantly and directly associated to plasma fT3 (the higher fT3, the better  systolic function). Vice-versa,  left ventricular mass was inversely associated to plasma fT3 (the lower fT3 the higher LVM).

Slide 15

Now, to explore the hypothesis that fT3, inflammation and hypoalbuminemia, that are  well established risk factors for left ventricular hypertrophy, are  in the same causal pathway leading to cardiomiopathy, we modelled the association between fT3, left ventricular mass and function in statistical models of increasing complexity and separately tested the effect of CRP, IL-6 and serum albumin on the regression coefficient of fT3.

Slide 16

Here in the ordinate scale, you see the beta regression coefficient that indicates the strength of the association between fT3 and mwFS. As previously  shown,  the crude correlation coefficient between fT3 and mwFS was 0.25. Adjustment of the beta coefficient of the fT3-mwFS relationship for Framingham risk factors and for anti-hypertensive therapy decreased this coefficient  from 0.25 to 0.17. The inclusion of  risk factors peculiar to ESRD into the model produced  a further small reduction  but the association  was  still significant. However,  adjustment  for CRP, IL-6 or, even more critically, for serum albumin   abrogated the  fT3-mwFS relationship.

Slide 17

Then we performed the same analysis with LVMI. You might remember that the correlation was  a negative one, the lowe fT3 the higher LVM, the crude correlation coefficient was -0.35. This coefficient decreased from -0.35 to -0.20 after  adjustment for Framingham risk factors and for anti-hypertensive therapy  and  showed an additional  small decrease  after adjustment  for risk factors peculiar to ESRD  which   was  still at a significant level. Again, data  adjustment  for CRP,  IL-6 or for albumin  made the correlation coefficient of fT3  weaker, the loss of statistical significance of the relationship between fT3 and LVMI being particularly pronounced  in the full model including serum albumin. In the aggregate these results suggest that low plasma fT3 is associated with Left ventricular dysfunction and LVH in ESRD patients. These associations appear largely mediated by inflammation and low fT3 may be implicated in the adverse cardiac effects of inflammation in patients with ESRD.

Slide 18

Now, the last point. Free-T3 and survival.

Slide 19

To have  a further insight on the potential importance of T3 in ESRD  we prospectively tested the hypothesis that it predicts death in these  patients. We studied a cohort of 200 HD patients and the follow up was of about 4 years. During the follow-up 102 patients died and as you can observe those patients who died had significantly lower plasma fT3 than those patients who survived.

Slide 20

In a Kaplan-Meyer analysis fT3 was associated with death. In fact  when we divided patients into three tertiles on the basis of individual values of plasma  fT3,  patients in the tertile with the lowest value of fT3, those  with the more severe thyroid illness,  had  the worst survival.

Slide 21

On multivariate Cox regression analysis, plasma fT3 remained a significant predictor of all-cause death. Indeed in a fully adjusted model including potential confounders such as age, sex, ADMA, and so on,  the strength of the association between plasma fT3 and death remained strong and significant : the higher plasma fT3 the lower the risk of death. Interestingly, in the multiple Cox model including plasma fT3, serum IL-6 failed to significantly predict  survival and it emerged as a significant predictor of death  only in a model excluding plasma fT3 suggesting that low fT3 and high IL-6 are in the same causal pathway leading to death in ESRD patients and that  fT3 mediates part of the adverse effects of inflammation in these patients.

Slide 22

LV dysfunction and LVH concur to the high mortality and CV complications of dialysis patients. Several risk factors and the inflammation/malnutrition syndrome are strongly associated to low fT3 which is the most precocious alteration signalling thyroid dysfunction. 

Slide 23

In conclusion, fT3 plasma levels are depressed in dialysis patients and are inversely associated to inflammation markers, such as IL-6 and CRP. Inflammation acutely interferes with thyroid function in patients with CKD and this interference is reversible.
Sustained inflammatory processes and low fT3 may induce severe cardiovascular damage and are important prognostic factors for survival in the ESRD population.
Low fT3 may not be an innocent adaptation to the inflammatory state of ESRD patients. Thank you.