Back to search

National and Kapodistrian University of Athens

Η τεκμηρίωση ενδείξεων και της αποτελεσματικότητας της συνεχούς αιμοδιαδιήθησης στον ασθενή της ΜΕΘ

Abstract

dc:description

The initiation of renal replacement therapy (RRT) has become a substantial therapeutic tool for critically ill patients and may play an important role in determination of their further outcome. However, ideal circumstances and indications, for RRT initiation have not been yet defined. Serum creatinine (sCr) failed to discriminate early structural damage and novel kidney markers specific for early diagnosis of RRT associated with acute kidney injury (AKI) cannot be incorporated into clinical practice yet.Our study provides information about the use of a combination of novel and traditional kidney markers for RRT prediction, taking into account that biomarkers have different profiles after injury, within time, and their predictive ability may vary with time and phase of injury. The primary outcome was the prediction of RRT need in ICU population, by using the combination of (conventional and novel) renal and clinical markers. The secondary outcome was the assessment of biomarker and microcirculatory alterations during RRT. For that reason, 106 patients (68 men/38 women) were included in the study and 25 patients were monitored for the first 24hours of RRT. Data collection included uNGAL, sCr and sCysC on ICU admission, before RRT initiation (H0), at six (H6) and 24 hours (H24) during RRT. Moreover, knowing that the microcirculation is involved in the pathophysiology of AKI, it was recorded during the 24 -hour hemodiafiltration at the same timepoints with biomarker sampling. The predictive accuracy of renal and clinical markers was evaluated and compared based on area under the receiver operating characteristic (ROC) curves. Further, the earlier timepoint where the markers presented their maximum area under the curve (AUC) has been assessed. At this specific timepoint (found at ten days after ICU admission) the optimum biomarker -cutoffs were determined. Then, multivariate cox regression analyses were conducted to assess the performance of the predictive models, while combining biomarkers with clinical factors. All studied biomarkers (uNGAL- nNGAL, sCysC and sCr on ICU admission), were significant independent predictors of RRT need (p<0.0001) -maximum follow up thirty ICU days-. sCysC showed the best discrimination of RRT need with an AUC of 0.765 (95% confidence interval, CI=0.67-0.86) compared to uNGAL (AUC=0.697, 95% CI=0.59-0.80), nNGAL (AUC=0.710, 95% CI=0.60-0.81) and sCr (AUC=0.756, 95%CI=0.66-0.85) on admission. Further, APACHE II score (AUC=0.723, 95%CI=0.63-0.82 was significant independent predictor of RRT need. The combination of nNGAL and sCr or sCysC was found as the best predictor of RRT initiation within ten days of ICU hospitalization (Wald test= 24.32 p<0.0001,AIC value: 315.97, AUC-ROC=0.801).When patients with RIFLE-AKI on ICU admission were excluded, the combination of sCysC and APACHE II score proved better (Wald test=11.9, p=0.003,AIC value: 179.22 AUC-ROC=0.776). uNGAL and sCysC values on ICU entry were higher among patients receiving RRT compared to those who did not and they increased significantly before RRT initiation. During the 24-hour RRT, the microcirculatory parameters {StO2 (76.5±12.5 vs75 ± 11 vs 70 ±16, p=0.04) και EF[7 (9) vs 1.55 (1.56) vs 1.23(2.17), p=0.02} decreased significantly, in parallel with sCysC (2.7 ± 0.8 vs 1.2 ± 0.6 vs 1.8 ± 0.8, p<0.0001), sCr (2.5 ± 1.5 vs 2 ± 1 vs 1.7 ± 0.9, p <0.0001), hemoglobin (9.5 ± 2 vs 9 ± 2 vs8 ± 2, p=0.004) and peripheral temperature [36.5 (0.8) vs 36 (0.9) vs 36.2(1.7), p=0.02]. In contrast, lactate levels increased [1.4 (0.9) vs1.8 (2.2) vs 2.2 (1.5), p= 0.015], and uNGAL did not change [968 (1398) vs1162.5 (1188) vs1145 (1526), p=0.9]. Finally, endothelial function alterations correlated with sCysC reduction during the 24 hour RRT (pearson, p=0.013 r= 0.464) and diabetics as well as patients receving RRT postdilution showed correlation of ΔStO2 with dose(ΔStO2 =8.1 ± 17%, mean dose 21 ± 5 ml/Kg/h, r=-0.52, p=0.03 και r=-0.8, p=0.01 respectively).Conclusively, the combination of sCr or sCysC and nNGAL was better predictor for RRT initiation than sCr, sCysC and nNGAL separately, over the first ten ICU days. When RRT prediction restricted to patients without established AKI on ICU entry, the combination of sCysC and APACHE II score was superior. During 24 -hour RRT, only sCysC, not uNGAL, presented significant changes. Microcirculatory alterations correlated with dose in patients with diabetes and those treated with hemodiafiltration on postdilution mode.

Degree

thesis:*
Grantor dc:publisher
National and Kapodistrian University of Athens
Year dc:date
2014

Author and committee

dc:creator, dc:contributor.*
Authors dc:creator
  • Pipili, Chrysoula
  • Πιπιλή, Χρυσούλα

Subjects

dc:subject × 17

Rights

Language dc:language
gre

Identifiers

dc:identifier.*
Identifier
10.12681/eadd/40236

Chain of custody

source
Harvested from
Greek National Archive of PhD Theses
Base URL
phdtheses.ekt.gr/eadd_oai/request
Last updated
2026-11-18
Source record
OAI-PMH GetRecord
citation

Pipili, Chrysoula; Πιπιλή, Χρυσούλα. Η τεκμηρίωση ενδείξεων και της αποτελεσματικότητας της συνεχούς αιμοδιαδιήθησης στον ασθενή της ΜΕΘ. National and Kapodistrian University of Athens, 2014. http://hdl.handle.net/10442/hedi/40236