OBJECTIVETo evaluate the use of clinical nutritional support in critical SARS patients, and the relationship between blood glucose levels/insulin administration amount and outcome.

METHODSTwenty-one SARS patients who reached the standard of Ministry of Health's "critical level" were transferred into our ICU in an average of 11 days after onset and enrolled in this clinical trial. All patients underwent respiratory support and clinical nutrition support as scheduled. For about 60 kg patient per day 3347.2 kJ(800 kcal), 36 g protein, and 125 g carbohydrate was given intravenously; 4184 kJ(1000 kcal), 38 g protein, and 125 g carbohydrate was provided by enteral route. MCT/LCT as fat resource shared 50% calories intake. All patients received similar doses of intravenous Methylprednisolone(about 200 mg/d). Blood glucose, serum albumin, blood lymphocyte counts, and serum alanine transminase (ALT) were checked on the first admission day in ICU and on the 12th day after nutrition therapy was started. Insulin was started to pump in to maintain the blood glucose levels between 4.44-7.78 mmol/L (80-140 mg/dl) when the levels exceeded normal range.

RESULTSUpon admission into ICU, all patients had poor nutrients intake for an average of 11 days and 16 patients (76.2%) were diagnosed as malnutrition. Parenteral and enteral nutrition therapy were then offered for an average of 12 days. On the 12th day, the serum albumin increased [(28.5 +/- 2.2)] g/L vs (37.0 +/- 4.1) g/L] (P = 0.0001) and so did the lymphocytes count [(0.74 +/- 0.47)] x 10(9)/L vs (1.22 +/- 0.73) x 10(9)/L] (P = 0.02). The blood glucose maintained at lower level in the surviving patients when compared with those who died [(9.5 +/- 2.3) mmol/L vs (6.3 +/- 1.8) mmol/L] [(196 +/- 70) mg/dl vs (110 +/- 21) mg/dl] (P = 0.0002), and the abnormally high ALT levels presented in some of the patients decreased but not significantly (81.0% vs 57.1%) (P = 0.18). In order to keep blood glucose within the range 4.44-7.78 mmol/L (80-140 mg/dl), only 18.8% of the surviving patients needed insulin intervention as opposed to 80.0% of those who died (P = 0.03). The amount of insulin used in the surviving group was significant lower than that in the group who died [(24 +/- 2) IU/d vs (72 +/- 9) IU/d] (P = 0.01).

CONCLUSIONSEleven days after SARS onset, most of the critical patients presented with malnutrition. Some improved nutrition related parameters may be associated with clinical nutritional support. The surviving patients required less insulin when compared to those who died. 80.0% of the patients who died need insulin versus only 18.8% of the surviving patients. Due to the difficulty of SARS management, this study was not a randomized controlled clinical trial. More clinical trials will be needed for checking the results of this investigation.