To investigate the related factors of serum carnitine deficiency in critical ill patients, and the influence of its deficiency on the length of hospital stay. A prospective study was conducted. Critical ill patients with acute physiology and chronic health evaluation II (APACHEII) score>12 admitted to Department of Critical Care Medicine of the First Affiliated Hospital of Sun Yat-sen University from March 2013 to September 2013 were enrolled. Serum carnitine concentration and indexes of organ function were determined, and the tolerance of enteral nutrition within 5 days, the length of hospital stay, the length of intensive care unit (ICU) stay, and the hospital mortality were recorded. The relationship between serum carnitine and indexes mentioned above was analyzed. Thirty critically ill patients were enrolled. Serum carnitine concentration was very low in all critically ill patients, i.e. (8.92 ± 5.05) μmol/L (normal reference value at 43.5 μmol/L) at hospital admission. Serum carnitine concentration in patients with APACHEII score>23 (7 cases) was significantly lower than that in those with APACHEII score 12-23 (23 cases, μmol/L: 5.33 ± 1.72 vs. 10.02 ± 5.24, t=2.300, P=0.001). Serum carnitine concentration in patients with serum total bilirubin(TBil)>19 μmol/L (9 cases) was significantly lower than that in those with TBil≤19 μmol/L (21 cases, μmol/L: 5.54 ± 2.70 vs. 9.84 ± 5.08, t=2.750, P=0.014). Serum carnitine concentration was negatively correlated with the APACHEII score and the TBil (r=-0.387, P=0.035; r=-0.346, P=0.048). During the 5-day observation period, enteral feeding amount [(5 134 ± 1 173) mL] was positively correlated with serum carnitine concentration(r=0.430, P=0.022). In 30 critical patients, the incidence of abdominal distension was 40.0% (12/30), and the serum carnitine concentration of patients with abdominal distension was lower compared with that of patients without abdominal distension (μmol/L: 7.83 ± 4.98 vs. 9.12 ± 5.35, t=0.707, P=0.383). The incidence of diarrhea was 26.7% (8/30), and the serum carnitine concentration of diarrhea patients was lower compared with that of patients without diarrhea (μmol/L: 8.27 ± 5.78 vs. 9.73 ± 4.78, t=0.607, P=0.576). The mean length of hospital stay was (34.72 ± 16.66) days. The serum carnitine concentrations in patients with hospital stay ≥ 45 days (8 cases) were lower compared with those in those <45 days (22 cases, μmol/L: 5.71 ± 3.23 vs. 9.95 ± 5.26, t=1.627, P=0.020). No correlation was found between serum carnitine concentrations and the hospital stay (r=-0.165, P=0.385). The length of ICU stay was (18.60 ± 10.72) days. Serum carnitine concentration in patients with the length of ICU stay>7 days (27 cases) was slightly lower than that in those with the length of ICU stay ≤ 7 days (3 cases, μmol/L: 8.44 ± 5.00 vs. 13.24 ± 3.65, t=1.610, P=0.119). No correlation was found between serum carnitine concentrations and the length of ICU stay (r=-0.019, P= 0.293). In-hospital mortality was 26.67% (8/30). No significant difference in serum carnitine concentrations was found between the death group and the survival group (μmol/L: 12.24 ± 6.52 vs. 7.72 ± 3.91, t=-1.846, P=0.098). No correlation was found between serum carnitine concentrations and in-hospital mortality (r=0.340, P=0.066). Carnitine deficiency is significant in critically ill patients, and it is correlated with disease severity and serum TBil. The total amount of lenteral feeding was lower, and hospital stay was prolonged in critically ill patients with low serum carnitine level.