Survival outcome in the nasal cannula group (n = 29) (A) and in the Venturi mask group (n = 17) (B), as described by stroke subtypes. AF indicates atrial fibrillation; PACI, partial anterior circulation infarction; and TACI, total anterior circulation infarction.
Chiu EHH, Liu C, Tan T, Chang K. Venturi Mask Adjuvant Oxygen Therapy in Severe Acute Ischemic Stroke. Arch Neurol. 2006;63(5):741-744. doi:10.1001/archneur.63.5.741
The effect of oxygen therapy in acute ischemic stroke remains undetermined.
To investigate the feasibility of eubaric hyperoxia therapy by Venturi mask (VM) in a group of patients who experienced a severe acute ischemic stroke.
Patients experiencing a first-ever large middle cerebral artery infarction were recruited within 48 hours after stroke. Patients were subdivided to undergo therapy with a VM with a fraction of inspired oxygen of 40% or with a nasal cannula. A large middle cerebral artery infarction was defined as a large low-attenuation area of more than one third of the middle cerebral artery territory on brain images. Stroke severity was evaluated by the National Institutes of Health Stroke Scale.
Seventeen patients were enrolled in the VM group and 29 in the nasal cannula group. All the demographic and clinical characteristics were equally distributed initially. The mean initial National Institutes of Health Stroke Scale score was 20.5 and 18.9 in the VM and nasal cannula groups, respectively. Atrial fibrillation was found in 11 (65%) patients in the VM and 17 (59%) patients in the nasal cannula groups. The VM therapy was initiated within 13.7 (range, 3.0-41.5) hours after stroke and the duration was 132.9 (range, 48.0-168.5) hours. In-hospital mortality was 1 (6%) in the VM group and 7 (24%) in the nasal cannula group (P=.12). In the VM group, there were fewer incidences of fever (4 [24%] vs 15 [52%]; P=.06), pneumonia (1 [6%] vs 6 [21%]; P=.18), and respiratory failure (3 [18%] vs 8 [28%]; P=.45), but a higher incidence of bedsores (3 [18%] vs 2 [7%]; P=.29).
By using VM therapy with a fraction of inspired oxygen of 40%, there might be less mortality and comorbidities in treated patients who experienced a severe acute ischemic stroke. Further randomized confirmatory studies should explore the decreased mortality in patients who experience a severe acute ischemic stroke, especially in those with a large middle cerebral artery infarction who undergo VM therapy with a fraction of inspired oxygen of 40%.
After stroke, supplemental oxygen is advised if there is evidence of hypoxia. However, data are lacking to support general use of supplemental oxygen in patients who experience an acute ischemic stroke. Hyperbaric oxygen (HBO) therapy has been studied for decades in patients who experience an acute ischemic stroke. The potential benefit of HBO might include increased oxygen delivery, decreased cerebral edema, and decreased lipid peroxidation.1 But the effectiveness of HBO therapy in humans is undetermined.2
Other than HBO therapy, eubaric (normobaric) hyperoxia treatment after acute stroke has been investigated recently. Eubaric hyperoxia treatment significantly increased penumbral PO2 during ischemia and increased penumbral PO2 during reperfusion.3 Experimental studies revealed that eubaric hyperoxia might protect the cerebral cortex4 and improve the neuropathological outcome.5 However, 1 atm of 100% oxygen for the first 24 hours was not recommended for nonhypoxic patients who experienced minor or moderate strokes.6
Compared with HBO therapy, eubaric hyperoxia can be applied with a Venturi mask (VM) (Figure 1), which is a simple and readily available device.7 The device delivered the most precise inspired oxygen fraction. Thus, we sought to investigate the feasibility and explore the therapeutic effects of eubaric hyperoxia by using a VM in a group of patients with a large middle cerebral artery (MCA) infarction, aiming to generate a hypothesis for future randomized trials.
This study involved 46 consecutive patients who experienced a first-ever severe ischemic stroke and were admitted within 48 hours after stroke onset at Changhua Christian Hospital between January 1, 2000, and July 31, 2001. Patients were subdivided into 2 groups: 17 received supplemental oxygen therapy by a VM with a fraction of inspired oxygen of 40% and 29 declined to participate in the program but agreed to participate as control subjects by using a nasal cannula, 2 L/min. The inclusion criteria were as follows: (1) first-ever ischemic stroke, (2) infarction in the MCA territory presenting less than 48 hours after onset, (3) National Institutes of Health Stroke Scale score of 12, (4) modified Rankin Scale score no less than 4, and (5) brain computed tomographic images suggesting that the infarction area was more than one third of the MCA territory. The exclusion criteria were as follows: (1) any known respiratory distress before registry period noted, such as obstructive pulmonary disease and obstructive sleep apnea; (2) rapidly improving neurological deficits; (3) brain images that did not reveal the infarction area was more than one third of the MCA territory; (4) medically unstable, due to previously known endocrinological dysfunctions, hematological abnormalities, autoimmune diseases, or malignancies; (5) pregnancy; (6) seizure at onset of stroke; (7) received thrombolytic therapy within 3 months; and (8) inability to obtain informed consent.
The oxygen devices would be changed or their use discontinued when the patients began ventilation; also, there was no need for supplemental oxygen, which was left at the discretion of the investigators. Stroke severity was assessed by the National Institutes of Health Stroke Scale.
There were no restrictions to concomitant treatments, such as an intravenous infusion of isotonic sodium chloride solution, hyperosmotic agents, and oral antiplatelet drugs. Patients who underwent craniectomy would not be included in this study. All fever episodes had been treated with empirical antibiotics.
Stroke subtypes were classified according to the Oxfordshire Community Stroke Project criteria.8 The specific factors at baseline were defined as in the Copenhagen Stroke Study.9 We compared the baseline characteristics and the outcomes, including survival and hospital length of stay (LOS) by Pearson χ2 or Fisher exact test if applicable. All values were reported as median (range) or mean (standard deviation), and tests were 2-tailed and differences considered to be statistically significant at P<.05. Data were analyzed using a commercially available software program (SPSS version 10.0 for Windows; SPSS Inc, Chicago, Ill).
Among 1013 consecutive patients who experienced a stroke and were admitted over a period of 18 months, 103 had a large MCA infarction. Only 46 of these patients with a large MCA infarction who had a first-ever ischemic stroke and were admitted within 48 hours after stroke onset were recruited into the analysis. The mean age of the patients was 79 years (median, 81 years; range, 49-93 years). Demographic characteristics are shown in Table 1. Atrial fibrillation (AF) was found in 28 (61%) of the 46 patients. There were no baseline differences between the 2 groups, except a slightly higher diastolic blood pressure in the nasal cannula group. The mean initial National Institutes of Health Stroke Scale score was 21 in the VM group and 18 in the nasal cannula group.
The VM therapy was initiated within 13.7 hours after stroke onset (median, 13 hours; range, 3.0-41.5 hours). The duration of VM use was 132.9 hours (median, 96 hours; range, 48.0-168.5 hours). All complications and vital status are shown in Table 2.
The mortality of the VM and nasal cannula groups was 1 (6%) and 7 (24%) (this percentage is different from the percentages totaled in the figure because of rounding) (P=.12) (Figure 2). One patient in the VM group died 4 days after stroke onset. Seven patients in the nasal cannula group died within 1 week after stroke.
The incidence of pneumonia was 1 (6%) in the VM group vs 6 (21%) in the nasal cannula group. There were fever episodes in 4 (24%) patients in the VM group and 15 (52%) patients in the nasal cannula group (P=.06). Asymptomatic petechial hemorrhages were seen in 2 patients in the VM group.
The overall LOS was 17 days (median, 13 days; range, 3-65 days): 17.2 days (median, 13 days; range, 4-58 days) in the VM group and 21.1 days (median, 13 days; range, 3-65 days) in the nasal cannula group. In patients with the total anterior circulation infarction subtype, the LOS was 17.9 days in the VM group vs 28.1 days in the nasal cannula group. Among the 11 patients with AF in the VM group, there was 1 death among those with the total anterior circulation infarction subtype and 0 deaths among those with the partial anterior circulation infarction subtype. Among the 17 patients with AF in the nasal cannula group, there were 5 deaths among those with the total anterior circulation infarction subtype and 1 death among those with the partial anterior circulation infarction subtype.
This small study suggested that VM therapy with a fraction of inspired oxygen of 40% might be beneficial in select patients who experienced a severe acute stroke; there was a trend in lowering the mortality, decreasing comorbidities, and a shortened hospital LOS compared with the nasal cannula group.
The reasons for the possible beneficial effects of eubaric hyperoxia in patients who experience an acute stroke are not yet clear. A recent study suggested eubaric hyperoxia treatment might result in an improvement in biochemical markers and reductions in the lactate-glucose and lactate-pyruvate ratios. Also, intracranial pressure was reduced during and after hyperoxia treatment.10 Although another report11 demonstrated a transient improvement of clinical deficits and magnetic resonance imaging abnormalities in select patients who experienced an acute ischemic stroke, further studies are warranted to investigate the safety and efficacy of hyperoxia as a stroke therapy.
In our observation, the incidence and duration of fever episodes were fewer and shorter in the VM group, which might be one of the reasons for the decreased comorbidity. Physiologically, fever increases the metabolic rate and oxygen consumption of the brain, which possibly make stroke outcome worse. Therefore, therapy with VM might have indirect beneficial effects of decreasing comorbidity after stroke.
Patients who have experienced both stroke and AF are at a higher risk of death. In our study, older female patients with AF seemed to have the worst outcomes. Not surprisingly, patients with AF had a National Institutes of Health Stroke Scale score plateau of 21 points compared with 18 points for those without AF (P=.05). Although the association of AF with immediate-phase mortality has not yet been confirmed,12 there was a higher mortality in the nasal cannula group with AF. Among the 6 mortalities with AF in the nasal cannula group, 3 patients had coronary artery disease. Venturi mask therapy might have a better effect in patients who have had a severe stroke and heart disease.
Obstructive sleep apnea might be a cause of ischemic stroke.13,14 However, we excluded those who had identified apnea syndromes or obstructive pulmonary diseases from the beginning. But, we were unable to identify patients with obstructive sleep apnea after stroke in this study. Blood gas data were imperfectly collected during investigation, which might need to be incorporated in future studies to alleviate the fear of inadequately maintaining oxygen.
In the study area, the median LOS of acute first-ever ischemic stroke was 6 to 25 days, which was the most important factor for the cost of short-term hospitalization.15,16 Because the results suggested survivors who had received VM therapy had an LOS that was shorter by 10 days (17.9 vs 28.1 days) when the 2 groups of patients with the total anterior circulation infarction subtype were compared, LOS might be an outcome that needs to be verified for this therapy in the future.
Patients with a large MCA infarction tend to have a poor prognosis. Moreover, after thrombolysis has been applied for acute ischemic stroke in most parts of the world, patients who have had a severe stroke are not supposed to undergo therapy. Thus, any measures to minimize mortality or comorbidities in these patients might have impact for stroke care. Our results supported the study of the beneficial effects of VM therapy with a fraction of inspired oxygen of40% in severe acute ischemic stroke. Future randomized trials are anticipated.
Correspondence: Ku-Chou Chang, MD, Department of Neurology, Chang Gung Memorial Hospital, 123 Tapei Rd, Niao Sung Hsiang, Kaohsiung County 833, Taiwan (firstname.lastname@example.org).
Accepted for Publication: December 19, 2005.
Author Contributions:Study concept and design: Chiu, Liu, and Chang. Acquisition of data: Chiu, Tan, and Chang. Analysis and interpretation of data: Chiu, Liu, and Chang. Drafting of the manuscript: Chiu, Liu, and Chang. Critical revision of the manuscript for important intellectual content: Chiu, Tan, and Chang. Statistical analysis: Liu. Administrative, technical, and material support: Chiu, Liu, Tan, and Chang. Study supervision: Chang.
Acknowledgment: We thank Shih-Lin Wu, MD, Mu-Chan Sun, MD, Zhen-Li Whu, MD, and Jane-Lin Kao and colleagues in the Department of Neurology, Changhua Christian Hospital, Changhua, Taiwan, for their advice and help in this study.