60
acutely stressed patients with a SIRS (n = 44) had significantly higher plasma con-
centrations of CGA than controls (115 μg/L (68–202.8) vs 40 μg/L (35–52.5),
p < 0.001). In addition, the septic group (sepsis, severe sepsis and septic shock) had
significantly higher CGA levels than patients with SIRS but without infections
(138.5 μg/L (65–222.3) vs 110 μg/L (81–143), p < 0.01). Interestingly, CGA plasma
levels positively correlated with biomarkers of inflammation (C-reactive protein,
procalcitonin and white cell counts) (Zhang et al. 2008 ).
In a multicentre analysis of VS-I in the plasma of critically ill patients (Tota et al.
2010 ), the population includes 481 patients and 13 healthy staff. Plasma VS-I con-
centrations were systematically higher in patients than in controls (p < 0.01). In
univariate analysis, only severe SIRS patients [5.77 ng/ml], severe sepsis [6.07 ng/
ml] and septic shock patients [6.21 ng/ml] displayed significantly higher plasma
VS-I concentrations when compared to no SIRS, no infection patients [3.56 ng/ml,
p < 0.01] and sepsis patients [3.6 ng/ml, p < 0.01]. Finally, there was a significant
increase in all patients demonstrating circulatory failure when compared to those
without circulatory failure [5.93 vs. 3.58 ng/ml, p < 0.001] and also in non-survivors
vs. survivors [5.75 vs. 3.70 ng/ml, p < 0.001], but there was no significant difference
in patients whether infection was present or not [4.41 vs. 3.88 ng/ml, p = 0.547]. As
shown in Fig. 6 , from admission to the 60th hour of admission within the ICU,
plasma VS-I concentrations were systematically higher in patients than in controls
(p < 0.01). We did not find differences of kinetics when comparing survivors and
non-survivors, or infected patients and non-infected patients.
The results of this study show for the first time that an increase in the VS-I con-
centration was present on admission in non-selected critically ill patients and that a
single measurement of this protein was associated with poor outcome at day 28 after
admission, without having to specify a primary diagnosis.
Studying the VS-I/CGA ratios, we have established that in physiological con-
ditions (in healthy controls), an average of 8% of the CGA is processed into VS-I,
whereas in critically ill patients this ratio was lower (around 5%). Despite this
absence of statistical difference among healthy controls and the entire cohort of
patients, surviving patients had in average a significantly lower VS-I/CGA ratio
than healthy controls (p < 0.05). Despite increased CGA levels after acute illness,
we only measured half of such a ratio in patients. This suggests that the cleavage
of CGA is limited under the pathological conditions of multiple organ failure.
Different hypothetical mechanisms can be proposed to explain this phenomenon.
First, CGA may aggregate (Yoo 1992 ) and this may hide the cleavage site located
at position 76–77 on CGA. Then, many enzymes (both from host and from patho-
gens), including metallo-proteases are overexpressed in sepsis (Yazdan-Ashoori
et al. 2011 ) and are candidates to degrade CGA and VS-I (Lauhio et al. 2011 ). In
addition, after acute stress, disseminated intravascular coagulation often occurs
(Lee and Downey 2000 ), which is a source of plasmin release that may interfere
with the cleavage of CGA (Parmer et al. 2000 ). Furthermore, VS-I accumulation
could result from an increase in the production of CGA and VS-I by nervous,
endocrine and also immune cells activated as a result of the initial stressing injury
(Wortsman et al. 1984 ).
M.-H. Metz-Boutigue and F. Schneider