Background Although severe encephalopathy has been proposed as a possible contraindication to the use ofnoninvasive positive-pressure ventilation(NPPV),increasing clinical reports showed it was effective in patients withimpaired consciousness and even coma secondary to acute respiratory failure,especially hypercapnic acute respiratoryfailure(HARF).To further evaluate the effectiveness and safety of NPPV for severe hypercapnic encephalopathy,aprospective case-control study was conducted at a university respiratory intensive care unit(RICU)in patients with acuteexacerbation of chronic obstructive pulmonary disease(AECOPD)during the past 3 years.Methods Forty-three of 68 consecutive AECOPD patients requiring ventilatory support for HARF were divided into 2groups,which were carefully matched for age,sex,COPD course,tobacco use and previous hospitalization history,according to the severity of encephalopathy,22 patients with Glasgow coma scale(GCS)<10 served as group A and 21with GCS>10 as group B.Results Compared with group B,group A had a higher level of baseline arterial partial CO2 pressure((102±27)mmHgvs(74±17)mmHg,P<0.01),lower levels of GCS(7.5±1.9 vs 12.2±1.8,P<0.01),arterial pH value(7.18±0.06 vs7.28±0.07,P<0.01)and partial O_2 pressure/fraction of inspired O_2 ratio(168±39 vs 189±33,P<0.05).The NPPV successrate and hospital mortality were 73%(16/22)and 14%(3/22)respectively in group A,which were comparable to those ingroup B(68%(15/21)and 14%(3/21)respectively,all P>0.05),but group A needed an average of 7 cmH_2O higher ofmaximal pressure support during NPPV,and 4,4 and 7 days longer of NPPV time,RICU stay and hospital stayrespectively than group B(P<0.05 or P<0.01).NPPV therapy failed in 12 patients(6 in each group)because of excessiveairway secretions(7 patients),hemodynamic instability(2),worsening of dyspnea and deterioration of gas exchange(2),and gastric content aspiration(1).Conclusions Selected patients with severe hypercapnic encephalopathy secondary to HARF can be treated aseffectively and safely with NPPV as awake patients with HARF due to AECOPD;a trial of NPPV should be instituted toreduce the need of endotracheal intubation in patients with severe hypercapnic encephalopathy who are otherwise goodcandidates for NPPV due to AECOPD. Background Currently severe encephalopathy has been proposed as a possible contraindication to the use of noninvasive positive-pressure ventilation (NPPV), increasing clinical reports showed it was effective in patients with impaired consciousness and even coma secondary to acute respiratory failure, especially hypercapnic acute respiratory failure (HARF) . To further evaluate the effectiveness and safety of NPPV for severe hypercapnic encephalopathy, aprospective case-control study was conducted at a university respiratory intensive care unit (RICU) in patients with acuteexacerbation of chronic obstructive pulmonary disease (AECOPD) during the past 3 years. Methods Forty-three of 68 consecutive AECOPD patients requiring ventilatory support for HARF were divided into 2 groups, which were carefully matched for age, sex, COPD course, tobacco use and previous hospitalization history, according to the severity of encephalopathy, 22 patients with Glasgow coma scale (GCS) <10 served as group A and 21 with GCS> 10 as g roup B. Results compared with group B, group A had a higher level of baseline arterial partial CO2 pressure ((102 ± 27) mmHg vs (74 ± 17) mmHg, P <0.01), lower levels of GCS ± 1.8, P <0.01), arterial pH value (7.18 ± 0.06 vs 7.28 ± 0.07, P <0.01) and partial O 2 pressure / fraction of inspired O 2 ratio (168 ± 39 vs 189 ± 33, P <0.05) NPPV successrate and hospital mortality were 73% (16/22) and 14% (3/22) respectively in group A, which were comparable to those ingroup B (68% (15/21) and 14% (3/21) respectively , all P> 0.05), but group A needed an average of 7 cmH 2 O higher of maximal pressure support during NPPV, and 4,4 and 7 days longer of NPPV time, RICU stay and hospital stay in group than group B (P <0.05 or P < 0.01). NPPPV therapy failed in 12 patients (6 in each group) because of excessiveairway secretions (7 patients), hemodynamic instability (2), worsening of dyspnea and deterioration of gas exchange (2), and gastric content aspiration (1). Conclusions Selected patients with severe hypercapnic encephalopathy seco ndary to HARF can be treated aseffectively and safely with NPPV as awake patients with HARF due to AECOPD; a trial of NPPV should be instituted toreduce the need of endotracheal intubation in patients with severe hypercapnic encephalopathy who are otherwise goodcandidates for NPPV due to AECOPD.