OBJECTIVE： To investigate risk factors of multidrug-resistant organisms （MDROs） infection in elderly patients of ICU， and to provide reference for formulation and implementation of MDROs prevention and control measures. ?METHODS： A total of 146 elderly patients were selected from ICU of our hospital during Dec. 2013-Jun. 2016. Throat swab， sputum swab and anal swab specimens （1 copy， respectively） were collected to conduct active screening of MRSA and ESBLs-producing Enterobacteriaceae. Risk factors of MDROs infection， pathogen distribution and drug resistance were analyzed. RESULTS： Among samples of 146 patients， there were 34 MRSA positive samples in throat swab with positive rate of 23.3％； there were 30 MRSA positive samples in sputum swab with positive rate of 20.5％； there were 99 ESBLs-producing bacteria positive samples in anal swab （containing 50 ESBLs-producing Escherichia coli positive samples and 49 ESBLs-producing Klebsiella pneumoniae positive samples） with positive rate of 67.8％. ?The positive rate of throat swab MRSA screening was not correlated with patient’s gender， age， tracheal intubation or mechanical ventilation （P＞0.05）， but it was related with hospitalization time in ICU （P＜0.05）. ?The positive rate of sputum swab MRSA screening was not correlated with patient’s gender， tracheal intubation or mechanical ventilation； the positive rate of anal swab ESBLs-producing bacteria screening were not related with patient’s gender（P＞0.05）. But they were related with age and hospitalization time in ICU （P＜0.05）. Compared with negative patients， there was no statistical significance in the times of fiberoptic bronchoscopy in throat/sputum swab MRSA screening positive patients （P＞0.05）. The times of enema， the times of bladder irrigation， the times of urethral catheterization and the duration of indwelling catheter in anal swab ESBLs-producing bacteria screening positive patients were significantly more or longer than negative patients， with statistical significance （P＜0.05）. Binary Logistic regression analysis showed that hospitalization time in ICU was risk factor of positive active screening of throat swab in elderly patients of ICU[OR＝1.119， 95％CI（1.071，1.385），P＝0.021]； age was risk factor of positive active screening of sputum swab[OR＝1.893， 95％CI（1.232，4.042），P＝0.032]； age and hospitalization time in ICU were risk factors of positive active screening of anal swab [OR were 1.046， 1.022， 95％CI were （1.005， 1.088） （1.006， 3.283）， P were 0.027，0.031]. A total of 163 strains of MDROs were detected， among which there were 64 strains of MRSA， 50 strains of ESBLs-producing E. coli and 49 strains of ESBLs-producing K. pneumoniae. They were generally highly resistant to compound preparation containing enzyme inhibitors. CONCLUSIONS： The results of MDROs active screening in elderly patients of ICU are related with age， hospitalization time in ICU， the times of enema， the times of bladder irrigation， the times of urethral catheterization and the duration of indwelling catheter. Age and hospitalization time in ICU were risk factors of MDROs infection. The pathogens are mainly ESBLs-producing Enterobacteriaceae， and drug resistance is severe. For elderly critical patients with MDROs infection， clinical prevention and intervention measures should be taken to prevent and control the prevalence and spread of MDROs in ICU.