Aneurysmal subarachnoid hemorrhage (SAH) accounts for a significant proportion of stroke-related mortality cases. The overall mortality rate among SAH cases has been reported to be over 30%, and approximately 10% to 20% of survivors remain functionally dependent despite intensive neurological care (1, 2). Several extensive studies have been performed to improve intensive neurological care in SAH patients (3–6).
Although oxygen management is necessary for SAH patients in the acute stage, hyperoxia is thought to have harmful effects. Several previous studies have shown an association between hyperoxia and poor outcomes in resuscitated patients after cardiac arrest, critically ill patients in the intensive care unit (ICU), patients with stroke, and patients with ST-segment elevation myocardial infarction (7–10).
Two previous studies investigated the association between hyperoxia and neurological outcomes in SAH patients. One study reported that hyperoxia, which was defined as arterial partial pressure of oxygen (PaO2) ≥173 mm Hg, was independently associated with poor outcomes (11). However, the other study reported that early moderate hyperoxia, which was defined as 24 h time-weighted arterial oxygen tension average (TWA-PaO2) >150 mm Hg, might not increase the risk of poor outcomes in mechanically ventilated SAH patients (12). Therefore, the effects of hyperoxia on neurological outcomes in SAH patients remain unknown. We hypothesized that hyperoxia leads to worse neurological outcomes in a certain group of SAH patients.
The purpose of this study was to examine the association of hyperoxia during the first 24 h in ICU with unfavorable neurological outcomes in SAH patients.
PATIENTS AND METHODS
Study design and setting
This single-center, retrospective cohort study was performed at Kagawa University Hospital, which is a 613-bed teaching institution with an eight-bed ICU managed by a neurointensivist. The hospital medical records were reviewed after obtaining approval from the institutional review board (IRB) (approval number: H29-090) and were assessed in accordance with the ethical standards established in the 1964 Declaration of Helsinki and its later amendments. A waiver of the requirement for individual subject consent for research was approved by IRB. If the patients did not want to participate in the current research, they could request to opt-out.
Study participants and inclusion criteria
We included patients aged ≥18 years, who were consecutively admitted to ICU between January 1, 2009 and April 30, 2018, for a confirmed diagnosis of SAH and were mechanically ventilated after ICU admission. The exclusion criteria were a history of trauma; presence of acute lung injury/acute respiratory distress syndrome or severe heart failure requiring positive pressure ventilation on admission, which directly affects oxygenation in SAH patients; and patients who were provided only comfort care.
General management of SAH in ICU
All patients were managed in accordance with the Guidelines for the Management of Aneurysmal Subarachnoid Hemorrhage by the American Heart Association/American Stroke Association (13). In addition to general intensive care, all patients were monitored for clinical deterioration or development of cerebral infarction associated with delayed cerebral ischemia (DCI). DCI was defined using previously published criteria as either the development of new focal neurological signs or deterioration in the level of consciousness thought to be due to the presence of ischemia, after excluding other possible causes of worsening or the appearance of new infarcts caused by vasospasm on computed tomography or magnetic resonance imaging (14, 15). Fluid management was targeted to maintain euvolemia. Induction of hypertension and hemodilution and administration of mannitol or hypertonic saline solution were not performed. For the general management of SAH in mechanically ventilated patients, hyperventilation was avoided with the use of sedatives, analgesics, and antipyretics, as appropriate. All patients were mechanically ventilated after ICU admission and were weaned off mechanical ventilation and allowed to breathe spontaneously or extubated after stabilization and completion of the initial treatment to secure the aneurysm with surgical clipping or endovascular coiling. Minimum amounts of sedatives, such as propofol, midazolam, and dexmedetomidine, which were necessary to prevent ventilator dyssynchrony and patient discomfort, were used. Analgesics, including acetaminophen, nonsteroidal anti-inflammatory medications, and fentanyl were administered as required. Fever was treated aggressively with acetaminophen, nonsteroidal anti-inflammatory medications, or cooling devices. Active maintenance of normothermia was not routinely performed.
Arterial blood gas analyses among SAH patients in ICU
Arterial blood gas (ABG) analyses, including PaO2, were routinely performed on admission and every 6 h during the first 2 weeks in ICU in our hospital. Additional measurements were performed by critical care physicians as needed. We collected the results of all ABG analyses performed during the first 24 h in ICU.
The following data were collected: age, sex, Hunt and Kosnik (H&K) grade, treatment approach (coil or clip), number of PaO2 measurements, PaO2 levels, minimum and maximum PaO2 levels during the first 24 h in ICU, modified Rankin scale (mRS) score at hospital discharge, rate of DCI, duration of mechanical ventilation, duration of ICU stay, duration of hospital stay, and hospital mortality. H&K grade is a useful grading scale for predicting the prognosis of SAH, which is well entrenched in the literature on SAH. In our study, we used the following grading scale: asymptomatic or minimal headache and slight nuchal rigidity; moderate-to-severe headache, nuchal rigidity, no neurological deficit other than cranial nerve palsy; drowsy, confusion, or mild focal deficit; stupor, moderate-to-severe hemiparesis, possibly early decerebrate rigidity and vegetative disturbances; and deep coma, decerebrate rigidity, moribund appearance (16, 17).
Definitions of hyperoxia
We defined hypoxia as PaO2 < 60 mm Hg, normoxia as PaO2 60 mm Hg to 120 mm Hg, and hyperoxia as PaO2 >120 mm Hg. Although there is no formal definition of hyperoxia, the above-mentioned PaO2 cutoff value was used in accordance with the cutoff value used in other previous studies that examined ICU hyperoxia exposure in mechanically ventilated patients (18–20). On the basis of the findings of a preceding study, we further grouped normoxia and hyperoxia together and categorized them into four hyperoxia subgroups as follows: normoxia (maximum PaO2 during the first 24 h in ICU, 60 mm Hg–120 mmHg); mild hyperoxia (maximum PaO2 during the first 24 h in ICU, 121 mm Hg–200 mm Hg); moderate hyperoxia (maximum PaO2 during the first 24 h in ICU, 201 mm Hg–300 mm Hg); and severe hyperoxia (maximum PaO2 during the first 24 h in ICU, >300 mm Hg) (20). We used these cutoff values to closely investigate the association of hyperoxia during the first 24 h in ICU with outcomes, as there is no consensus about specific cutoff points of hyperoxia in SAH patients.
The primary outcome was the association of hyperoxia during the first 24 h in ICU with unfavorable neurological outcomes, which were assessed using the mRS at hospital discharge (21). The mRS is a measure of global disability and comprises the following seven outcome categories: no symptoms at all, no significant disability, slight disability, moderate disability, moderately severe disability, severe disability, and death. We evaluated all SAH patients in real time using the mRS at hospital discharge and assessed their medical records. The neurological outcome was defined as unfavorable when the mRS score was 3 to 6 and as favorable when the score was 0 to 2. The secondary outcome was the association of hyperoxia during the first 24 h in ICU with the presence of DCI.
To estimate the sample size, we assumed the odds ratio (OR) between hyperoxia and neurological outcome in SAH as 2.5 (11). The study needed 156 patients to achieve 80% power with a 2-sided α level of 0.05 (22, 23).
Patients were divided into four groups according to their PaO2 levels (hyperoxia subgroups). Demographic factors and baseline characteristics were summarized using descriptive statistics. The groups were compared using Student t test or the Mann–Whitney U test, as appropriate. Categorical comparisons were performed using Fisher exact test. Univariate and multivariate analyses were conducted to identify the independent predictors of unfavorable neurological outcomes. The covariates of age (>65 years), sex (female), H&K grade, treatment approach (coil or clip), and the four hyperoxia subgroups were included in the multivariate analysis. Relative risk (RR) was calculated by the formula including OR (24). Statistical analyses were performed using EZR (Saitama Medical Center, Jichi Medical University, Saitama, Japan), which is a graphical user interface for R (The R Foundation for Statistical Computing, Vienna, Austria); more precisely, it is a modified version of R commander designed to add statistical functions that are frequently used in biostatistics (25). A two-sided P value of <0.05 was considered statistically significant for all analyses. Missing data were not replaced or estimated.
Baseline characteristics of the study population
The study included 196 patients (mean age, 62.7 years; 63 men; Table 1). The median number of PaO2 measurements during the first 24 h in ICU was 5 per patient. Unfavorable neurological outcomes were observed in 45.9% of the patients. Hyperoxia (defined as PaO2 > 120 mm Hg) during the first 24 h in ICU was observed in 93.4% of the patients.
Baseline characteristics of the study population across hyperoxia subgroups
On univariate analysis (Table 2), there was no significant difference in age (P = 0.63) and H&K grade (P = 0.39). However, patients who were categorized into the severe hyperoxia group accounted for 61.5% of the unfavorable neurological outcomes.
Association between hyperoxia subgroups and unfavorable neurological outcomes
As the level of hyperoxia increased, the proportion of patients with unfavorable neurological outcomes increased; however, this was not statistically significant among the hyperoxia subgroups (Fig. 1). Patients who were categorized into the severe hyperoxia group accounted for 61.5% of the unfavorable neurological outcomes, whereas patients who were categorized into the normoxia group accounted for only 23.1% of the unfavorable neurological outcomes (Fig. 1).
Predictors of unfavorable neurological outcomes in SAH patients
On multivariate analysis (Table 3, Supplementary Table 1, https://links.lww.com/SHK/A786), hyperoxia subgroups were not significantly associated with unfavorable neurological outcomes (RR, 1.38; 95% CI, 0.99–1.83; P = 0.06).
On multivariate analysis (Table 3, Supplementary Table 2, https://links.lww.com/SHK/A786), unfavorable neurological outcomes were significantly associated with hyperoxia subgroups for H&K grades I–III (RR, 1.84; 95% CI, 1.10–2.94; P = 0.02). The associations between hyperoxia subgroups and unfavorable neurological outcomes for H&K grades I to III are shown in Supplementary Figure 1 (https://links.lww.com/SHK/A787). Patients who were categorized into the severe hyperoxia group accounted for 55.6% of the unfavorable neurological outcomes.
Baseline characteristics of the study population across hyperoxia subgroups (H&K grades I–III)
Baseline characteristics of the patients with H&K grades I to III across hyperoxia subgroups are shown in Supplementary Table 3 (https://links.lww.com/SHK/A786). Unfavorable neurological outcomes were observed in 35.4% of these patients. For H&K grades I to III, patients who were categorized into the severe hyperoxia group accounted for 55.6% of the unfavorable neurological outcomes.
Association between PaO2 levels and DCI
DCI was observed in 13.8% of the patients. The duration from admission to the occurrence of DCI was 8 [7–9] days. The presence of DCI was confirmed during the first 2 weeks in ICU. Comparisons of clinical characteristics between patients with DCI and those without DCI are presented in Supplementary Table 4 (https://links.lww.com/SHK/A786). On univariate analysis, there were no significant differences in age, sex, treatment approach, and hypoxia or hyperoxia during the first 24 h in ICU. However, the durations of ICU and hospital stay were significantly longer in patients with DCI than in those without DCI. In addition, DCI was associated with unfavorable neurological outcomes.
The present study found no significant association between hyperoxia during the first 24 h in ICU and unfavorable neurological outcomes among all SAH patients. However, hyperoxia during the first 24 h in ICU was associated with unfavorable neurological outcomes among SAH patients with H&K grades I to III.
One previous report, which showed a significant association between hyperoxia and neurological outcomes among SAH patients in multiple logistic regression analysis (not univariate analysis), had the primary objective of investigating the effects of hyperoxia on DCI (11). Thus, this report included ABG analyses until the point when each patient developed DCI or until post-bleed day 6 among patients who did not develop DCI. On the other hand, our study included all ABG analyses performed during the first 24 h in ICU. The period of the first 24 h is long enough for hyperoxia to be harmful and short enough to investigate comparatively brief exposure to hyperoxia, as this is a critical period for SAH patients. Additionally, excessive administration of oxygen after admission to ICU has been reported to be common in critically ill patients (9, 18).
A previous report on the association between hyperoxia and neurological outcomes in SAH patients targeted the first 24 h of mechanical ventilation (12). This previous study concluded that early moderate hyperoxia might not affect outcomes in mechanically ventilated SAH patients. However, the study could not sufficiently account for severe hyperoxia because of few occurrences of severe hyperoxia (defined as PaO2 > 300 mm Hg). Therefore, the effect of hyperoxia in SAH patients was unclear and controversial. Alternatively, the present study included SAH patients having mild-to-severe hyperoxia. We investigated the association of hyperoxia with unfavorable neurological outcomes in acute-phase SAH and identified the subgroups in which hyperoxia influenced neurological outcomes.
Mechanism of the effect of early hyperoxia on neurological outcomes and DCI
A negative effect of hyperoxia on outcomes has been reported in various patient populations. The mechanism of the negative effect is well understood in systematic circulation and the brain. In fact, hyperoxia has been reported to induce vasoconstriction and decrease cardiac output, which can reduce blood flow and oxygen transport (26, 27). Hyperoxia also drives the formation of reactive oxygen species at sites of tissue injury, which can cause neuronal inflammation in the brain. It is considered a major cause of secondary brain injury, and an injured brain is vulnerable to low oxygen levels (28–30). Furthermore, hyperoxia is known to be associated with a decrease in cerebral blood flow (31). Additionally, oxidative stress can trigger processes that can lead to cerebral aneurysm development and rupture through inflammation, activation of matrix metalloproteinases, and lipid peroxidation (32).
In our study, early hyperoxia was not associated with neurological outcomes in the entire group of SAH patients. However, in subgroup analysis, a significant association between neurological outcomes and early hyperoxia was noted in SAH patients with H&K grades I to III. The association between hyperoxia and neurological outcomes in SAH patients has been investigated (11, 12). However, there are some differences between these previous studies and our study with regard to the period of ABG analyses as the subject or hyperoxia severity. To our knowledge, the present analysis is the first to evaluate a more direct functional outcome and to demonstrate the effect of early hyperoxia on unfavorable neurological outcomes in SAH patients with H&K grades I to III. We assumed that in severe cases (H&K grade IV–V subgroups), primary brain damage is severe, and the effect of early hyperoxia would not be evaluated precisely. It has been reported that typical ICU patients at risk for hyperoxia might be less critically ill ICU patients (33). Therefore, attention should be paid to hyperoxia even in patients who are not seriously ill.
Our current findings are somewhat consistent with the results of two previous studies (11, 12), although there are differences with respect to the PaO2 cutoff value. On the other hand, with regard to DCI, one previous report showed an association between DCI and hyperoxia from ICU admission until the development of DCI. However, our study investigated early hyperoxia mainly during the first 24 h in ICU. In addition, the study included a small number of patients with DCI. Accordingly, there was no significant association between early hyperoxia and DCI in our study.
Although our study design was not controlled intentionally, the findings of our study collectively suggested that early hyperoxia above a certain level was associated with unfavorable neurological outcomes in SAH patients, especially those with H&K grades I to III. Recent previous studies reported the potentially harmful effects of hyperoxia in ICU. Nevertheless, there is no consensus guideline on the optimal PaO2 target level for patients in ICU. Patients may benefit from careful monitoring or tighter control of PaO2, especially in the early stage in ICU, and at least avoiding PaO2 levels above 300 mm Hg appears reasonable.
This study had several limitations. First, there was potential selection bias because this was a retrospective cohort study conducted at a single center. Moreover, uncontrolled confounding factors may have been present. Second, the neurological outcomes of patients after discharge were not assessed. Third, we could not definitively exclude the possible impact of differences in the patient population, such as those associated with comorbidities, on neurological outcomes. Fourth, the oxygen dose or the specific mechanical ventilator setting, including the fraction of inspired oxygen and positive end-expiratory pressure during each ABG analysis, was not sufficiently considered. Fifth, the sample size in this study was relatively small. Finally, because the current study examined the limited association between early hyperoxia and neurological outcomes in SAH patients, we did not determine the effect of regulation of early hyperoxia. Furthermore, as details of causes of early hyperoxia such as prehospital oxygen therapy and oxygen concentration on mechanical ventilation were not obtained because of the nature of the retrospective study, we could not suggest definitive idea regarding oxygen therapy for clinical use. Therefore, further prospective study including RCTs will be required to confirm the effect of early hyperopia on neurological outcome in SAH patients.
Hyperoxia frequently occurs during the first 24 h in ICU among SAH patients. Early hyperoxia was not associated with unfavorable neurological outcomes in overall SAH patients, but it was associated with unfavorable neurological outcomes in those with H&K grades I to III. A further study is needed to provide evidence on the effects of hyperoxia or identify the optimal PaO2 target level in SAH patients.
The authors are grateful to all physicians and nurses at the study site for their crucial contributions to the successful completion of this study.
1. Hop JW, Rinkel GJ, Algra A, van Gijn J. Case-fatality rates and functional outcome after subarachnoid hemorrhage: a systematic review. Stroke
2. Grasso G, Alafaci C, Macdonald RL. Management of aneurysmal subarachnoid hemorrhage: state of the art and future perspectives. Surg Neurol Int
3. Okazaki T, Hifumi T, Kawakita K, Shishido H, Ogawa D, Okauchi M, Shindo A, Kawanishi M, Tamiya T, Kuroda Y. Blood glucose variability: a strong independent predictor of neurological outcomes in aneurysmal subarachnoid hemorrhage. J Intensive Care Med
4. Okazaki T, Hifumi T, Kawakita K, Shishido H, Ogawa D, Okauchi M, Shindo A, Kawanishi M, Inoue S, Tamiya T, et al. Serial blood lactate measurements and its prognostic significance in intensive care unit management of aneurysmal subarachnoid hemorrhage patients. J Crit Care
5. Okazaki T, Hifumi T, Kawakita K, Shishido H, Ogawa D, Okauchi M, Shindo A, Kawanishi M, Tamiya T, Kuroda Y. Target serum sodium levels during intensive care unit management of aneurysmal subarachnoid hemorrhage. Shock
6. Egawa S, Hifumi T, Kawakita K, Okauchi M, Shindo A, Kawanishi M, Tamiya T, Kuroda Y. Impact of neurointensivist-managed intensive care unit implementation on patient outcomes after aneurysmal subarachnoid hemorrhage. J Crit Care
7. Wang CH, Huang CH, Chang WT, Tsai MS, Lu TC, Yu PH, Wang AY, Chen NC, Chen WJ. Association between early arterial blood gas tensions and neurological outcome in adult patients following in-hospital cardiac arrest. Resuscitation
8. Damiani E, Adrario E, Girardis M, Romano R, Pelaia P, Singer M, Donati A. Arterial hyperoxia and mortality in critically ill patients: a systematic review and meta-analysis. Crit Care
9. Rincon F, Kang J, Maltenfort M, Vibbert M, Urtecho J, Athar MK, Jallo J, Pineda CC, Tzeng D, McBride W, et al. Association between hyperoxia and mortality after stroke: a multicenter cohort study. Crit Care Med
10. Stub D, Smith K, Bernard S, Nehme Z, Stephenson M, Bray JE, Cameron P, Barger B, Ellims AH, Taylor AJ, et al. Air versus oxygen in ST-segment-elevation myocardial infarction. Circulation
11. Jeon SB, Choi HA, Badjatia N, Schmidt JM, Lantigua H, Claassen J, Connolly ES, Mayer SA, Lee K. Hyperoxia may be related to delayed cerebral ischemia and poor outcome after subarachnoid haemorrhage. J Neurol Neurosurg Psychiatry
12. Lang M, Raj R, Skrifvars MB, Koivisto T, Lehto H, Kivisaari R, von Und Zu Fraunberg M, Reinikainen M, Bendel S. Early Moderate hyperoxemia does not predict outcome after aneurysmal subarachnoid hemorrhage. Neurosurgery
13. Connolly ES Jr, Rabinstein AA, Carhuapoma JR, Derdeyn CP, Dion J, Higashida RT, Hoh BL, Kirkness CJ, Naidech AM, Ogilvy CS, et al. Guidelines for the management of aneurysmal subarachnoid hemorrhage: a guideline for healthcare professionals from the American Heart Association/american Stroke Association. Stroke
14. Frontera JA, Fernandez A, Schmidt JM, Claassen J, Wartenberg KE, Badjatia N, Connolly ES, Mayer SA. Defining vasospasm after subarachnoid hemorrhage: what is the most clinically relevant definition? Stroke
15. Williamson CA, Sheehan KM, Tipirneni R, Roark CD, Pandey AS, Thompson BG, Rajajee V. The Association between spontaneous hyperventilation, delayed cerebral ischemia, and poor neurological outcome in patients with subarachnoid hemorrhage. Neurocrit Care
16. Hunt WE, Kosnik EJ. Timing and perioperative care in intracranial aneurysm surgery. Clin Neurosurg
17. Rosen DS, Macdonald RL. Subarachnoid hemorrhage grading scales: a systematic review. Neurocrit Care
18. Eastwood G, Bellomo R, Bailey M, Taori G, Pilcher D, Young P, Beasley R. Arterial oxygen tension and mortality in mechanically ventilated patients. Intensive Care Med
19. Helmerhorst HJ, Arts DL, Schultz MJ, van der Voort PH, Abu-Hanna A, de Jonge E, van Westerloo DJ. Metrics of arterial hyperoxia and associated outcomes in critical care. Crit Care Med
20. Page D, Ablordeppey E, Wessman BT, Mohr NM, Trzeciak S, Kollef MH, Roberts BW, Fuller BM. Emergency department hyperoxia is associated with increased mortality in mechanically ventilated patients: a cohort study. Crit Care
21. Banks JL, Marotta CA. Outcomes validity and reliability of the modified Rankin scale: implications for stroke clinical trials: a literature review and synthesis. Stroke
22. Demidenko E. Sample size determination for logistic regression revisited. Stat Med
23. Demidenko E. Sample size and optimal design for logistic regression with binary interaction. Stat Med
24. Zhang J, Yu KF. What's the relative risk
? A method of correcting the odds ratio in cohort studies of common outcomes. JAMA
25. Kanda Y. Investigation of the freely available easy-to-use software ’EZR’ for medical statistics. Bone Marrow Transplant
26. Farquhar H, Weatherall M, Wijesinghe M, Perrin K, Ranchord A, Simmonds M, Beasley R. Systematic review of studies of the effect of hyperoxia on coronary blood flow. Am Heart J
27. Cornet AD, Kooter AJ, Peters MJ, Smulders YM. The potential harm of oxygen therapy in medical emergencies. Crit Care
28. Seder DB, Jagoda A, Riggs B. Emergency Neurological life support: airway, ventilation, and sedation. Neurocrit Care
23 (suppl 2):S5–S22, 2015.
29. Nakanishi H, Wu Z. Microglia-aging: roles of microglial lysosome- and mitochondria-derived reactive oxygen species in brain aging. Behav Brain Res
30. Robertson CL, Scafidi S, McKenna MC, Fiskum G. Mitochondrial mechanisms of cell death and neuroprotection in pediatric ischemic and traumatic brain injury. Exp Neurol
31. Floyd TF, Clark JM, Gelfand R, Detre JA, Ratcliffe S, Guvakov D, Lambertsen CJ, Eckenhoff RG. Independent cerebral vasoconstrictive effects of hyperoxia and accompanying arterial hypocapnia at 1 ATA. J Appl Physiol (1985)
32. Starke RM, Chalouhi N, Ali MS, Jabbour PM, Tjoumakaris SI, Gonzalez LF, Rosenwasser RH, Koch WJ, Dumont AS. The role of oxidative stress in cerebral aneurysm formation and rupture. Curr Neurovasc Res
33. Kraft F, Andel H, Gamper J, Markstaller K, Ullrich R, Klein KU. Incidence of hyperoxia and related in-hospital mortality in critically ill patients: a retrospective data analysis. Acta Anaesthesiol Scand