Fig. 3

Survival of patients with sepsis in quartile groups of plasma symmetric (SDMA) and asymmetric dimethylarginine (ADMA) concentration. Mortality curves were calculated for 28-day survival and three groups were compared: patients with SDMA levels (a) or ADMA levels (b) ≤ 25th percentile, between 25th and the 75th percentile (interquartile) and > 75th percentile. a The three groups differed significantly (P = 0.004): patients with SDMA levels > 1.34 μmol/L had the highest mortality and patients with SDMA levels ≤ 0.65 μmol/L the lowest mortality. b There was no difference in mortality between patients in the interquartile and ≤ 25th percentile (P = 0.715); however, patients with ADMA levels > 75th percentile had the highest mortality compared to other groups (P = 0.022). c Decision tree to identify patient risk. Out of 120 patients, 31 died (26%). First decision knot: (1) patients were identified as having intermediate risk, when SDMA levels were ≤ 1.34 μmol/L. This group was further risk stratified by SDMA levels; (2) low-risk patients had levels ≤ 0.65 and median-risk patients had levels between 0.65 and 1.34 μmol/L. Mortality increased to 43% when SDMA levels were > 1.34 μmol/L, indicating high risk. At this step ADMA levels may help to identify patients with the highest risk of not surviving sepsis; (3) plasma ADMA concentrations > 0.97 μmol/L were associated with a 61% mortality rate