Sie können Operatoren mit Ihrer Suchanfrage kombinieren, um diese noch präziser einzugrenzen. Klicken Sie auf den Suchoperator, um eine Erklärung seiner Funktionsweise anzuzeigen.
Findet Dokumente, in denen beide Begriffe in beliebiger Reihenfolge innerhalb von maximal n Worten zueinander stehen. Empfehlung: Wählen Sie zwischen 15 und 30 als maximale Wortanzahl (z.B. NEAR(hybrid, antrieb, 20)).
Findet Dokumente, in denen der Begriff in Wortvarianten vorkommt, wobei diese VOR, HINTER oder VOR und HINTER dem Suchbegriff anschließen können (z.B., leichtbau*, *leichtbau, *leichtbau*).
Volatile anesthetics have been suggested to exert neuroprotective effects in patients with subarachnoid hemorrhage caused by a ruptured cerebral aneurysm. However, their effects on functional outcomes remain unverified. We assessed the association between volatile anesthetics and functional outcomes in patients with subarachnoid hemorrhage.
Methods
Using data from the Japanese Diagnosis Procedure Combination inpatient database, patients with subarachnoid hemorrhage, aged = 18 years, and undergoing any procedures for aneurysm treatment were selected. Patients were categorized into those who received volatile anesthetics and those who did not. The primary outcome was a composite of in-hospital death or impaired functional outcome at discharge. The secondary outcomes included in-hospital mortality, postoperative cerebral infarction, postoperative acute hydrocephalus, tracheostomy, hospital stay, and total healthcare costs. After 1:1 propensity-score matching, a generalized linear model or linear model was applied for each outcome, with cluster-robust standard error adjustment. Interaction analysis was also conducted for the primary outcome and in-hospital mortality.
Results
Overall, 35,097 matched pairs were generated. No significant difference was noted in the primary outcome between the two groups (total intravenous anesthetics: 44.5%; volatile anesthetics: 44.1%; odds ratio 0.99, 95% confidence interval [CI] 0.93–1.06, p = 0.84). However, in-hospital mortality differed significantly between the groups (total intravenous anesthetics: 9.3%; volatile anesthetics: 8.7%; odds ratio 0.89, 95% CI 0.82–0.97, p < 0.01). Other secondary outcomes showed no significant group differences. Interaction analysis indicated that volatile anesthetics worsened outcomes among patients with impaired consciousness at admission.
Conclusions
Volatile anesthetic use was not associated with improved functional outcomes in patients with subarachnoid hemorrhage. In patients presenting with impaired consciousness, volatile anesthetics were associated with poorer outcomes, although this finding should be interpreted with caution, given the observational nature of the study.
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
BBB
Blood–brain barrier
CI
Confidence interval
ICD-10
International Classification of Diseases, Tenth Revision
JCS
Japan Coma Scale
SD
Standard deviation
Background
Subarachnoid hemorrhage is most commonly caused by the rupture of a cerebral arterial aneurysm, with high mortality rates, approximately 45% [1]. It is characterized by worsening cognitive and functional abilities, including activities of daily living. This impairment is one of the most critical problems affecting long-term outcomes in patients with subarachnoid hemorrhage. Several studies have reported that approximately 20–50% of patients experience cognitive deficits and reduced activities of daily living after subarachnoid hemorrhage [2, 3].
Subarachnoid hemorrhage-associated brain damage is caused by early brain injury and delayed cerebral ischemia. Early brain injury, which manifests within 72 h post-subarachnoid hemorrhage, is characterized by blood–brain barrier (BBB) disruption, cellular death, neuroinflammation, and brain edema and independently predicts adverse outcomes [4]. Delayed cerebral ischemia is also one of the most critical complications associated with subarachnoid hemorrhage, occurring 4–14 days after the initial subarachnoid hemorrhage caused by vasospasm and autoregulatory failure [5] and is related to focal neurological impairment. Several factors, including neuroinflammation and microvascular dysfunction, contribute to delayed cerebral ischemia and subsequent impairment in patients with subarachnoid hemorrhage [6, 7]. Both types of brain damage significantly impact patient mortality and neurological outcomes [8, 9]. Therefore, developing effective therapeutic and preventive strategies for early brain injury and delayed cerebral ischemia is crucial for improving patient prognoses. However, proven strategies for preventive treatment are limited [10‐12].
Anzeige
Volatile anesthetics, known for their neuroprotective effects, have been used in animal studies to investigate the effectiveness of preventing both early brain injury and delayed cerebral ischemia in subarachnoid hemorrhage models [13, 14]. Additionally, several studies have demonstrated the beneficial effects of volatile anesthetics in preventing delayed cerebral ischemia during aneurysm treatment [15, 16]. However, these studies only focused on the association between delayed cerebral ischemia and volatile anesthetics and did not verify the association between functional outcomes and volatile anesthetics. Furthermore, the sample size of each study was small, which limited the ability to determine whether the use of volatile anesthetics reduces impaired activities of daily living, a clinically significant outcome. However, prior studies were small and focused mainly on surrogate endpoints, such as delayed cerebral ischemia. Hence, the clinical relevance of volatile anesthetics remains uncertain, underscoring the need for large-scale real-world evaluation.
We hypothesized that volatile anesthetics would reduce the incidence of adverse clinical outcomes, including mortality and impaired functional status, in patients with subarachnoid hemorrhage. Therefore, the primary objective of the present study was to evaluate the association between the use of volatile anesthetics and a composite outcome of in-hospital mortality and functional impairment at discharge.
Methods
Study design and data source
This study utilized a retrospective cohort design based on routinely gathered data, following the REporting of studies Conducted using Observational Routinely collected health Data guidelines [17]. The Tohoku University Institutional Review Board approved the study (approval number 2022–1–444; August 2022). Due to the anonymized nature of data, the board waived the requirement for obtaining written informed consent.
The data for this study were obtained from the Japanese Diagnosis Procedure Combination (DPC) inpatient database, which includes discharge summaries and administrative claims from approximately 1100 acute-care hospitals across Japan. This database gathers annual data on approximately seven million patients, accounting for approximately half of all acute-care hospitalizations in the country. It records extensive information, including patient demographics, smoking history, body measurements, initial consciousness level, functional status at admission, use of ambulances, surgical details, and diagnoses categorized under the International Classification of Diseases, Tenth Revision (ICD-10) codes and presented in Japanese. Additionally, it includes details on medical and surgical procedures, medication prescriptions, drug administration practices, and patient outcomes upon discharge [18, 19]. The accuracy of diagnostic entries is crucial for reimbursement, requiring physicians to substantiate their diagnoses with concrete evidence to qualify for compensation [20].
Anzeige
Study population
Using Diagnosis Procedure Combination data from April 1, 2012 to March 31, 2022, we selected all patients aged = 18 years, admitted owing to subarachnoid hemorrhage (ICD-10 code: I60), and who underwent any surgical procedure for aneurysm treatment. These included aneurysm clipping (procedure codes: K176 or K177) and interventional radiology (procedure code: K178). We excluded patients who underwent surgical procedures 4 days after admission, those with missing Japan Coma Scale (JCS) data [21], which was recorded as the consciousness level at admission, those with missing composite outcomes as the primary outcome, and those with missing information concerning total healthcare costs. The DPC database includes neurological status exclusively assessed using JCS, and data on assessments using the Glasgow Coma Scale (GCS) is not available. Approximate correspondences reported previously indicate that JCS scores of 0, 1–3, 10–30, and 100–300 generally correspond to GCS scores of 15, 13–15, 9–12, and 3–7, respectively, whereas the most severe category, corresponding to a JCS score of 300, is generally regarded as equivalent to a GCS score of 3 [22]. The category with a JCS score of 1–3 includes patients with very mild impairment, many of whom still correspond to a GCS score of 14–15, which explains why both the categories with JCS scores of 0 and 1–3 encompass patients with preserved levels of consciousness [22].
Group allocation and outcomes
We categorized the patients into two groups: the total intravenous anesthetic group, serving as the comparator group, and the volatile anesthetic group, serving as the exposure group. From the billing code, we identified patients who received volatile anesthetics [23].
The primary outcome was the composite outcome of patient death and impaired functional outcomes at hospital discharge, as defined by the following parameters: (1) JCS score of 10–300, indicating unconsciousness or coma; (2) use of mechanical ventilation; (3) tracheostomy; and (4) impaired functional outcomes, as defined by a Barthel index (BI) of < 60 [24]. This composite outcome was used because these endpoints represent mutually exclusive but equally meaningful adverse outcomes after subarachnoid hemorrhage; combining them avoids survivor bias arising from missing functional status among patients who die during hospitalization. In the DPC system, the BI is a mandatory data item recorded at hospital discharge, typically assessed by the rehabilitation staff or nursing personnel, based on direct patient evaluation. The BI has been widely used in previous DPC-based studies [25, 26].
Secondary outcomes included in-hospital mortality; incidence of postoperative cerebral infarction (ICD-10 code: I63) and postoperative acute hydrocephalus (ICD-10 code: G91 or procedure code: K174); proportion of patients who underwent tracheostomy (procedure code; K386); length of hospital stay; and total healthcare costs. All costs were originally recorded in Japanese yen and then converted into U.S. dollars using an exchange rate of 1 USD = 150 JPY.
Data collection
We extracted patient demographic and clinical variables considered as potential confounding variables, including sex, age, body mass index category (< 18.5, 18.5–22.5, 22.5–25, 25–30, or = 30.0 kg.m-2 or missing value), smoking status, ambulance utilization, admission to a teaching hospital, fiscal year category (2012–2014, 2015–2017, and 2018–2021), Charlson comorbidity index [27], JCS score at admission [21], and the “difficulty in anesthesia” criteria. This is a Japan-specific, claim-based classification used for reimbursement adjustment and is not directly linked to the American Society of Anesthesiologists Physical Status (ASA-PS). Although some items conceptually overlap with conditions often categorized as ASA-PS = III, the system has not been validated against the ASA-PS classification. The detailed items are summarized in Supplementary Table S1 (see Additional file 1) [28]. The following surgical information was collected: type of surgery, location of the aneurysm, additional surgical procedures (continuous ventricular drainage or spinal drainage), and timing of surgery. We also extracted preoperative therapeutic information, including admission to the intensive care unit or high-dependency unit, catecholamine use, albumin use, respiratory support therapy, and the use of opioids or sedatives. Although standard subarachnoid hemorrhage severity grading scales (e.g., Hunt–Hess, World Federation of Neurosurgical Societies (WFNS), Fisher) are essential for accurately assessing baseline subarachnoid hemorrhage severity, such information, including imaging data, was not available in the DPC claims database [29‐31].
Statistical analysis
We used 1:1 propensity-score matching to control for confounding factors between the two groups [32]. We employed a multivariable logistic regression model to predict the propensity score of patients receiving volatile anesthesia on the day of surgery. All covariates shown in Table 1 were used to calculate the propensity score. One-to-one nearest-neighbor matching without replacement was performed for propensity-score matching, using a caliper width of 20% of the standard deviation (SD) of the propensity score. The distribution of propensity scores was illustrated before and after matching to facilitate visual understanding. An absolute standardized difference of < 10% was considered to indicate an acceptable balance between the two groups [33].
Table 1
Baseline characteristics of patients undergoing treatment for subarachnoid hemorrhage after propensity-score matching
Matched cohort
Total intravenous anesthesia
Volatile anesthesia
Variable
(n = 35,097)
(n = 35,097)
ASD
Age, years old, mean (SD)
63.4 (14.2)
63.3 (14.2)
0.9
Sex, female, n (%)
24,532 (69.9)
24,434 (69.6)
0.6
Smoking history, n (%)
1.5
Non-smoker
20,309 (57.9)
20,270 (57.8)
Current/past smoker
9,907 (28.2)
10,093 (28.8)
Unknown
4,881 (13.9)
4,734 (13.5)
Body mass index, kg/m2, n (%)
1.2
< 18.5
3,892 (11.1)
3,793 (10.8)
18.5–22.4
13,337 (38.0)
13,286 (37.9)
22.5–24.9
7,682 (21.9)
7,722 (22.0)
25.0–29.9
6,073 (17.3)
6,165 (17.6)
= 30.0
1,518 (4.3)
1,555 (4.4)
Unknown
2,595 (7.4)
2,576 (7.3)
Japan Coma Scale at admission, n (%)
2.8
0
9,143 (26.1)
9,512 (27.1)
1–3
9,657 (27.5)
9,322 (26.6)
10–30
7,927 (22.6)
7,983 (22.7)
100–300
8,370 (23.8)
8,280 (23.6)
Ambulance use, n (%)
29,703 (84.6)
29,558 (84.2)
1.1
Financial year, n (%)
8.7
2012–2014
9,779 (27.9)
10,121 (28.8)
2015–2017
11,700 (33.3)
12,784 (36.4)
2018–2021
13,618 (38.8)
12,192 (34.7)
Location of aneurysm, n (%)
Internal carotid artery
10,116 (28.8)
10,124 (28.8)
0.1
Middle cerebral artery
9,138 (26.0)
8,789 (25.0)
2.3
Anterior communicating artery
9,012 (25.7)
9,226 (26.3)
1.4
Posterior communicating artery
753 (2.1)
740 (2.1)
0.3
Basilar artery
1,325 (3.8)
1,348 (3.8)
0.3
Vertebral artery
2,022 (5.8)
2,049 (5.8)
0.3
Others
2,790 (7.9)
2,868 (8.2)
0.8
Type of surgery, n (%)
0.2
Interventional radiology
11,350 (32.3)
11,324 (32.3)
Clipping surgery
23,747 (67.7)
23,773 (67.7)
Additional surgical procedure at the time of surgery, n (%)
Continuous ventricular drainage
1,457 (4.2)
1,423 (4.1)
0.5
Spinal drainage
2,853 (8.1)
2,928 (8.3)
0.8
Timing of surgery, n (%)
1.4
Weekday
26,711 (76.1)
26,500 (75.5)
Weekend
8,386 (23.9)
8,597 (24.5)
Charlson comorbidity index, median (IQR)
0.0 [0.0, 1.0]
0.0 [0.0, 1.0]
0.0
Difficulty of anesthesia, n (%)
8,067 (23.0)
8,341 (23.8)
1.8
Admission to ICU or HDU before surgery, n (%)
4,154 (11.8)
4,188 (11.9)
0.3
Use of dopamine before surgery, n (%)
94 (0.3)
101 (0.3)
0.4
Use of dobutamine before surgery, n (%)
16 (0.0)
20 (0.1)
0.5
Use of noradrenalin before surgery, n (%)
68 (0.2)
85 (0.2)
1.0
CPR before surgery, n (%)
15 (0.0)
16 (0.0)
0.1
Renal replacement therapy before surgery, n (%)
8 (0.0)
12 (0.0)
0.7
Conventional oxygen therapy before surgery, n (%)
5,541 (15.8)
5,399 (15.4)
1.1
Non-invasive ventilation before surgery, n (%)
5 (0.0)
5 (0.0)
0.0
Invasive mechanical ventilation before surgery, n (%)
1,716 (4.9)
1,769 (5.0)
0.7
Use of fentanyl before surgery, n (%)
1,863 (5.3)
1,778 (5.1)
1.1
Use of midazolam before surgery, n (%)
1,951 (5.6)
1,954 (5.6)
0.0
Use of propofol before surgery, n (%)
4,024 (11.5)
3,983 (11.3)
0.4
Use of dexmedetomidine before surgery, n (%)
1,275 (3.6)
1,243 (3.5)
0.5
Use of thiopental before surgery, n (%)
30 (0.1)
32 (0.1)
0.2
Use of albumin on the day of surgery, n (%)
12 (0.0)
13 (0.0)
0.2
ASD absolute standardized difference, CPR cardiopulmonary resuscitation, HDU high-dependency unit, ICU intensive care unit, SD standard deviation
Utilizing the propensity score-matched cohort, a generalized linear model was employed for binary outcomes, including the primary outcome. Additionally, a linear model was used to analyze continuous outcomes. These models accounted for a comprehensive set of variables, encompassing patient demographics, clinical characteristics, and treatment details, as outlined in the ‘‘Data collection’’ section.
To account for potential intrahospital correlations and improve the robustness of our estimates, we used cluster-robust standard errors, clustering by hospital code [34]. We performed this analysis using the vcovCL function in R (https://sandwich.r-forge.r-project.org/reference/vcovCL.html), which computes cluster-robust standard errors for generalized linear models. This approach provided more reliable confidence intervals (CIs) and p-values in the presence of clustering [35]. The odds ratios or differences and 95% CIs for each value were obtained according to the analyses.
Anzeige
For a sensitivity analysis, we conducted the same analysis using inverse probability of treatment weighting to account for potential confounding and estimate the average treatment effect [36]. The propensity score used in the main analysis was also applied in the sensitivity analysis. Stabilized weights were assigned to each individual to maintain the sample size of the original dataset, reduce variance inflation, and improve the precision of interval estimates. We employed a weighted generalized linear model using a stabilized average treatment effect weight to evaluate each outcome, with cluster-robust standard errors accounting for individual hospitals as clusters.
We conducted interaction analyses to identify specific subgroups where volatile anesthesia may confer benefits. We examined both the primary outcome and in-hospital mortality, evaluating the interaction terms between volatile anesthetic use and (1) age (as a continuous variable), (2) sex, and (3) neurological status at admission (categorized according to the JCS). Each interaction term was included in a multivariable logistic regression model, along with other covariates, using the propensity score-matched cohort. Cluster-robust standard errors according to the hospital code were used to account for within-hospital correlation; the results were summarized in a forest plot.
Continuous variables were reported as mean and SD or median and interquartile range, whereas categorical variables were presented as counts and percentages. Two-sided p-values < 0.05 were considered statistically significant. All analyses were conducted using R version 4.4.0 (2024–04–24).
Results
Patient flow chart
In this database, 94,785 patients were admitted due to subarachnoid hemorrhage. Based on the inclusion criteria, 88,178 patients were eligible for this study. We excluded 813 patients who lacked information on the JCS scores at admission, 701 without information on the primary outcome, and 3183 who lacked information on admission costs. Overall, 83,550 patients were included in this study (Fig. 1).
Patients’ characteristics and distribution of propensity score
Supplementary Table S2 (see Additional file 2) provides the baseline characteristics of the unmatched cohort, and Table 1 presents the baseline characteristics of the matched cohorts. In the unmatched cohort, the volatile anesthesia group exhibited higher proportions of middle cerebral artery aneurysms, utilized clipping surgery as the primary treatment more frequently, and difficulty with anesthesia, all of which showed a high absolute standardized difference. After propensity-score matching, we obtained 35,097 1:1 matched pairs without replacement. Supplementary Fig. S1 (see Additional file 3) illustrates the distribution of the propensity score, indicating similar patterns after matching. After matching, the baseline characteristics between the two groups were well balanced, with an absolute standardized difference of < 10%. The proportion of volatile anesthetics utilized during surgeries was as follows: for 27,925 (79.6%) patients, sevoflurane was used; for 7205 (20.5%), desflurane was used; and for 728 (2.1%), isoflurane was used; while for 761, two volatile anesthetics were used during surgery.
Primary and secondary outcomes
Table 2 presents the outcomes in the propensity score-matched cohort. In this cohort, the primary outcome was not significantly different between the two groups (total intravenous anesthetic group: 44.5% vs. volatile anesthetic group: 44.1%; odds ratio 0.99, 95% CI 0.93–1.06), p = 0.84). Regarding secondary outcomes, the in-hospital mortality rate was significantly different between the groups (total intravenous anesthetic group: 9.3% vs. volatile anesthetic group: 8.7%; odds ratio 0.89, 95% CI 0.8–0.95, p < 0.01). The incidence of postoperative cerebral infarction and postoperative hydrocephalus, proportion of patients who underwent tracheostomy, and length of hospital stay and total healthcare costs at discharge were not significantly different between the two groups.
Table 2
Propensity score-matched cohort outcomes
Total intravenous anesthetics
Volatile anesthetics
Odds ratio or difference
Outcomes
(n = 35,097)
(n = 35,097)
(95% CI)
p-value
Composite outcome, n (%)
15,608 (44.5)
15,486 (44.1)
0.99 (0.93–1.06)
0.837
In-hospital mortality, n (%)
3250 (9.3)
3055 (8.7)
0.89 (0.82–0.97)
0.008
Postoperative cerebral infarction, n (%)
5807 (16.5)
5763 (16.4)
0.98 (0.87–1.10)
0.706
Postoperative hydrocephalus, n (%)
6844 (19.5)
7114 (20.3)
1.05 (0.98–1.13)
0.157
Tracheostomy, n (%)
3106 (8.8)
3022 (8.6)
0.97 (0.87–1.07)
0.493
Length of hospital stay, days, median (IQR)
35.0 [24.0, 56.0]
36.0 [25.0, 56.0]
0.64 (- 1.17–2.44)
0.489
Total healthcare cost, Dollars, median (IQR)
28,532.2 [22608.9, 37,350.3]
28,788.3 [22992.6, 37,555.7]
522.3 (- 14.6–1059.1)
0.056
CI confidence interval, IQR interquartile range
Sensitivity analyses
Sensitivity analysis was also performed. Table 3 illustrates the comparison of patient backgrounds between the two groups after the inverse probability of treatment weighting, showing that the two groups were well-balanced. Table 4 demonstrates the results of the inverse probability of treatment weighting. The primary outcome was not significantly different (total intravenous anesthetic group: 44.9% vs. volatile anesthetic group: 44.3%; odds ratio 0.97, 95% CI 0.97–0.98, p = 0.40). In the sensitivity analysis, only the in-hospital mortality rate was significantly lower in the volatile anesthetic group (total intravenous anesthetic group: 9.6% vs. volatile anesthetic group: 8.5%; odds ratio 0.85, 95% CI 0.85–0.85, p < 0.01).
Table 3
Baseline characteristics of patients undergoing treatment for subarachnoid hemorrhage after inverse probability of treatment weighting
After inverse probability of treatment weighting
Variable
Total intravenous anesthetics
Volatile anesthetics
ASD
Age, years, mean (SD)
63.5 (14.3)
63.5 (14.3)
0.1
Sex, female, %
70.1
70.1
0.1
Smoking history, %
0.2
Non-smoker
58.0
58.0
Current/past smoker
27.8
27.8
Unknown
14.2
14.1
Body mass index, kg/m2, %
0.3
< 18.5
11.1
11.1
18.5–22.4
38.1
38.1
22.5–24.9
21.7
21.7
25.0–29.9
17.1
17.2
= 30.0
4.4
4.4
Unknown
7.6
7.5
Japan Coma Scale score at admission, %
0.4
0
25.7
25.8
1–3
27.2
27.3
10–30
22.4
22.4
100–300
24.7
24.5
Ambulance use, %
84.7
84.6
0.1
Financial year, %
0.5
2012–2014
27.7
27.5
2015–2017
33.5
33.6
2018–2021
38.8
38.9
Location of aneurysm, %
Internal carotid artery
29.4
29.4
0.0
Middle cerebral artery
24
24.1
0.2
Anterior communicating artery
25.7
25.6
0.1
Posterior communicating artery
2.1
2.2
0.2
Basilar artery
4.3
4.3
0.1
Vertebral artery
6.7
6.6
0.2
Others
7.9
7.9
0.2
Type of surgery, %
0.4
Interventional radiology
32.3
32.3
Clipping surgery
60.8
61.0
Additional surgical procedure at the time of surgery, %
Continuous ventricular drainage
4.4
4.4
0.0
Spinal drainage
9.0
9.0
0.1
Timing of surgery, %
0.1
Weekday
76.1
75.5
Weekend
23.9
24.5
Charlson comorbidity index, median (IQR)
0.0 [0.0, 1.0]
0.0 [0.0, 1.0]
0.1
Difficulty of anesthesia, %
22.3
22.5
0.4
Admission to ICU or HDU before surgery, %
11.9
11.9
0.2
Use of dopamine before surgery, %
0.3
0.3
0.1
Use of dobutamine before surgery, %
0.1
0.1
0.4
Use of noradrenaline before surgery, %
0.3
0.3
0.2
CPR before surgery, %
0.1
0.1
0.3
Renal replacement therapy before surgery, %
0.0
0.0
0.0
Conventional oxygen therapy before surgery, %
15.6
15.7
0.3
Non-invasive ventilation before surgery, %
0.0
0.0
0.2
Invasive mechanical ventilation before surgery, %
5.8
5.6
0.6
Use of fentanyl before surgery, %
6.1
6.0
0.3
Use of midazolam before surgery, %
5.7
5.7
0.1
Use of propofol before surgery, %
12.2
12.1
0.2
Use of dexmedetomidine before surgery, %
4.1
4.1
0.0
Use of thiopental before surgery, %
0.1
0.1
0.2
Use of albumin on the day of surgery, %
0.0
0.0
0.0
ASD absolute standardized difference, CPR cardiopulmonary resuscitation, HDU high-dependency unit, ICU intensive care unit, SD standard deviation
Table 4
Sensitivity analysis results
Total intravenous anesthetics
Volatile anesthetics
Odds ratio or difference
Outcomes
(95% CI)
p-value
Composite outcome, %
44.9
44.3
0.97 (0.92–1.03)
0.397
In-hospital mortality, %
9.6
8.5
0.85 (0.78–0.92)
< 0.001
Postoperative cerebral infarction, n (%)
16.6
16.3
0.98 (0.87–1.10)
0.736
Postoperative hydrocephalus, n (%)
19.4
20.1
1.05 (0.98–1.13)
0.153
Tracheostomy, n (%)
9.1
8.5
0.92 (0.83–1.02)
0.100
Length of stay, days, median (IQR)
35.0
[24.0, 55.0]
36.0
[25.0, 56.0]
0.48 (-1.25–2.22)
0.587
Total healthcare cost, Dollars, median (IQR)
28,585.7 (22,632.0, 37,374.0)
28,813.0 (23,046.7, 37,619.3)
495.5 (-27.1–1018.1)
0.063
CI, confidence interval, IQR interquartile range. Significant p-values (p < 0.05) are given in bold
Interaction analyses
Figure 2 illustrates the results of the interaction analyses between volatile anesthetic use and patients’ characteristics, including age, sex, and neurological status at admission. No significant interaction was observed with age (continuous) or sex in relation to either the primary composite outcome or in-hospital mortality. However, the neurological status at admission significantly interacted with the effect of volatile anesthetics. Compared with patients with normal consciousness (JCS = 0), those with impaired consciousness demonstrated higher odds of experiencing the primary composite outcome. The interaction between volatile anesthesia and consciousness level was significant in all impaired categories (JCS 1–3: odds ratio 1.21, 95% CI 1.06–1.38, p < 0.01; JCS 10–30: odds ratio 1.19, 95% CI 1.04–1.35, p < 0.01; and JCS 100–300: odds ratio 1.20, 95% CI 1.05–1.37, p < 0.01). For in-hospital mortality, significant interactions were observed in patients with moderate-to-severe impaired consciousness (JCS 10–30: odds ratio 1.29, 95% CI 1.01–1.64, p = 0.04; JCS 100–300: odds ratio 1.23, 95% CI 1.01–1.49, p = 0.04). Full results from the interaction analyses are provided in Supplementary Table S3 (see Additional file 4).
Fig. 2
Interaction analysis of patients’ characteristics on the association between volatile anesthetic use and outcomes after subarachnoid hemorrhage. CI, confidence interval; JCS, Japan Coma Scale
This nationwide retrospective observational study evaluated the efficacy of volatile anesthetic administration during treatment for subarachnoid hemorrhage in improving functional outcomes. We used propensity-score matching for the main analysis and inverse probability of treatment weighting for the sensitivity analysis to adjust for covariates. The use of volatile anesthetics was not associated with the composite outcomes of in-hospital mortality, impaired activities of daily living, unconsciousness, and tracheostomy. The sensitivity analysis revealed the same trend for each outcome. However, the in-hospital mortality rate was lower in the volatile anesthetic group in both the main and sensitivity analyses, suggesting a possible mortality reduction effect. Additional interaction analyses suggested that patients with preserved neurological status at admission could have benefited from using volatile anesthetics, whereas those with impaired consciousness did not experience any benefits. These findings indicate that, although volatile anesthetics may be associated with improved outcomes overall, their effects may be unfavorable in patients presenting with neurological deterioration. To the best of our knowledge, no previous nationwide database study has attempted to verify the effects of volatile anesthetics on the functional outcomes in patients undergoing treatment for subarachnoid hemorrhage.
Volatile anesthetics have several neuroprotective effects, including anti-inflammatory properties and a reduction in endothelin-1 levels, which can induce vasoconstriction [37]. In addition to animal experiments, several observational studies have reported possible delayed cerebral ischemia prevention effects when using volatile anesthetics [38]. However, these studies did not evaluate the association between volatile anesthetic use during subarachnoid hemorrhage surgery and the functional outcomes at discharge. In our study, we defined the primary outcome as clinically important findings, focusing on mortality rate and functional outcomes. Nevertheless, despite a sufficient sample size, we could not demonstrate the benefits of volatile anesthetics regarding the composite outcome. Our results suggest that the benefits observed in basic research do not fully translate to improved neurological function in patients.
In this study, the in-hospital mortality rate was lower in the volatile anesthetic group. The sensitivity analysis showed similar results. This may be due to the protective effects of volatile anesthetics against early brain injury. The pathophysiological mechanisms of early brain injury include initial mechanical injury, microcirculatory dysfunction, BBB disruption, neuroinflammation, and apoptosis [39]. Animal studies have demonstrated that volatile anesthetics can reduce neuroapoptosis, BBB disruption, and neuroinflammation [14, 40]. However, no studies have assessed whether volatile anesthetics reduce mortality in subarachnoid hemorrhage. Thus, our results may prompt further large-scale clinical trials. Theoretically, volatile anesthetic use during subarachnoid hemorrhage surgery may have a greater impact on early brain injury than on delayed cerebral ischemia because the administration of volatile anesthetics occurs earlier (< 72 h in our study cohort). Our results elucidate the potential benefits of volatile anesthetics in preventing early brain injury. Although a significant difference was observed, the absolute mortality difference was < 1%, suggesting limited clinical relevance. This small difference should be interpreted cautiously, given the observational nature of the study.
However, the use of volatile anesthetics should be cautiously approached in patients with impaired consciousness. Our interaction analyses revealed that the use of volatile anesthetics in patients with impaired consciousness (defined as a JCS = 10) was associated with an increased risk of both the primary composite outcome and in-hospital mortality. One possible explanation is that volatile anesthetics may exacerbate non-convulsive status epilepticus [41], which is one of the causes of unconsciousness in subarachnoid hemorrhage [42]. Another potential explanation is that impaired cerebral autoregulation in patients with reduced consciousness may alter the cerebrovascular response to volatile anesthetics, potentially leading to unfavorable hemodynamic or metabolic effects in vulnerable brain tissue [43, 44]. These mechanistic pathways remain unproven and should be interpreted with caution.
This study has some limitations. First, owing to the nature of the DPC database, we could not obtain data on several factors that affected the outcome, including computed tomography data, GCS scores, and subarachnoid hemorrhage severity based on the WFNS, Hunt–Hess, and Fisher grading scales [29‐31]. Thus, we used the JCS as an alternative to the GCS. We incorporated several parameters into the propensity score matching, including preoperative mechanical ventilation, sedative use, and opioid use before surgery, and also conducted subgroup analyses. However, the JCS itself is an incomplete substitute for the GCS because the scaling methodology is different, even though a previous study demonstrated the validity of the JCS as an alternative to the GCS [21]. Therefore, residual confounding may exist, owing to the nature of the database. In particular, the results observed among patients with a JCS = 10 should be interpreted with caution, as unmeasured differences in baseline subarachnoid hemorrhage severity may partly account for this finding. Second, the decision to use volatile anesthetics was made at the discretion of each anesthesiologist and facility, causing confounding by indication. Third, we could not assess the dose–outcome relationship. Although we obtained information on the amount of volatile anesthetic administered, we could not obtain data on the end-tidal volatile anesthetic concentration, as they are recorded in individual anesthetic records. End-tidal volatile anesthetic concentrations vary depending on the total gas flow and target concentration settings of the anesthetic machine, making it challenging to verify dose-dependent effects. In an animal experiment, an excessive sevoflurane concentration resulted in the loss of neuroprotective effects [45], indicating the necessity of an optimal dose setting. Fourth, data on the ASA-PS and detailed intraoperative physiological parameters, such as blood pressure, end-tidal CO2, temperature, and anesthesia depth (e.g., Bispectral Index or other processed electroencephalographic metrics) were not available in the DPC database. These unmeasured factors could have influenced the anesthetic requirements and neurological outcomes. Fifth, we did not assess the actual incidence of delayed cerebral ischemia because it is diagnosed using magnetic resonance angiography or angiography and is not captured by ICD-10 codes. Therefore, our study could not exclude the possible preventive effects of volatile anesthetics on delayed cerebral ischemia. Nevertheless, we primarily aimed to evaluate the efficacy of volatile anesthetics regarding functional outcomes.
Conclusions
Volatile anesthetic use was not associated with improved functional outcomes in patients undergoing treatment for subarachnoid hemorrhage. Although a small mortality difference was observed, its clinical significance remains uncertain. Interaction analysis suggested a possible increased risk among patients presenting with impaired consciousness. Further prospective randomized controlled trials are warranted to validate these findings.
Acknowledgements
None.
Declarations
Ethics approval and consent to participate
The Tohoku University Institutional Review Board approved the study (approval number 2022–1-444; August 2022). Due to the anonymized nature of data, the board waived the requirement for obtaining written informed consent.
Consent for publication
Not applicable.
Competing interests
The authors declare no competing interests.
Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Pobereskin LH (2001) Incidence and outcome of subarachnoid haemorrhage: a retrospective population based study. J Neurol Neurosurg Psychiatry 70:340–343PubMedPubMedCentralCrossRef
2.
Powell J, Kitchen N, Heslin J, Greenwood R (2002) Psychosocial outcomes at three and nine months after good neurological recovery from aneurysmal subarachnoid haemorrhage: predictors and prognosis. J Neurol Neurosurg Psychiatry 72:772–781PubMedPubMedCentralCrossRef
3.
Hackett ML, Anderson CS (2000) Health outcomes 1 year after subarachnoid hemorrhage: an international population-based study. Neurology 55:658–662PubMedCrossRef
4.
Sehba FA, Hou J, Pluta RM, Zhang JH (2012) The importance of early brain injury after subarachnoid hemorrhage. Prog Neurobiol 97:14–37PubMedPubMedCentralCrossRef
5.
Dodd WS, Laurent D, Dumont AS, Hasan DM, Jabbour PM, Starke RM et al (2021) Pathophysiology of delayed cerebral ischemia after subarachnoid hemorrhage: a review. J Am Heart Assoc 10:e021845PubMedPubMedCentralCrossRef
6.
Zheng VZ, Wong GKC (2017) Neuroinflammation responses after subarachnoid hemorrhage: a review. J Clin Neurosci 42:7–11PubMedCrossRef
7.
Yamada H, Kase Y, Okano Y, Kim D, Goto M, Takahashi S et al (2022) Subarachnoid hemorrhage triggers neuroinflammation of the entire cerebral cortex, leading to neuronal cell death. Inflamm Regen 42:61PubMedPubMedCentralCrossRef
8.
Nieuwkamp DJ, Setz LE, Algra A, Linn FHH, de Rooij NK, Rinkel GJE (2009) Changes in case fatality of aneurysmal subarachnoid haemorrhage over time, according to age, sex, and region: a meta-analysis. Lancet Neurol 8:635–642PubMedCrossRef
9.
Abdulazim A, Heilig M, Rinkel G, Etminan N (2023) Diagnosis of delayed cerebral ischemia in patients with aneurysmal subarachnoid hemorrhage and triggers for intervention. Neurocrit Care 39:311–319PubMedPubMedCentralCrossRef
10.
Allen GS, Ahn HS, Preziosi TJ et al (1983) Cerebral arterial spasm—a controlled trial of nimodipine in patients with subarachnoid hemorrhage. N Engl J Med 308:619–624PubMedCrossRef
11.
Pickard JD, Murray GD, Illingworth R, Shaw MD, Teasdale GM, Foy PM et al (1989) Effect of oral nimodipine on cerebral infarction and outcome after subarachnoid haemorrhage: British aneurysm nimodipine trial. BMJ 298:636–642PubMedPubMedCentralCrossRef
12.
Hoh BL, Ko NU, Amin-Hanjani S, Chou SH-Y, Cruz-Flores S, Dangayach NS et al (2023) guideline for the management of patients with aneurysmal subarachnoid hemorrhage: a guideline from the American Heart Association/American Stroke Association. Stroke 2023(54):e314–e370
13.
Jayaraman K, Liu M, Zipfel GJ, Athiraman U (2021) Sevoflurane and desflurane exposures following aneurysmal subarachnoid hemorrhage confer multifaceted protection against delayed cerebral ischemia. Biomedicines 9:820PubMedPubMedCentralCrossRef
14.
Beck-Schimmer B, Restin T, Muroi C, Roth Z’graggen B, Keller E, Schläpfer M (2020) Sevoflurane sedation attenuates early cerebral oedema formation through stabilisation of the adherens junction protein beta catenin in a model of subarachnoid haemorrhage: a randomised animal study. Eur J Anaesthesiol 37:402–412PubMedCrossRef
15.
Athiraman U, Aum D, Vellimana AK, Osbun JW, Dhar R, Tempelhoff R et al (2020) Evidence for a conditioning effect of inhalational anesthetics on angiographic vasospasm after aneurysmal subarachnoid hemorrhage. J Neurosurg 133:152–158PubMedCrossRef
16.
Athiraman U, Dhar R, Jayaraman K, Karanikolas M, Helsten D, Yuan J et al (2021) Conditioning effect of inhalational anesthetics on delayed cerebral ischemia after aneurysmal subarachnoid hemorrhage. Neurosurgery 88:394–401PubMedCrossRef
17.
Benchimol EI, Smeeth L, Guttmann A, Harron K, Hemkens LG, Moher D et al (2016) The REporting of studies Conducted using Observational Routinely-collected health Data (RECORD) statement. Z Evid Fortbild Qual Gesundhwes 115–6:33–48CrossRef
18.
Shigemi D, Aso S, Yasunaga H (2019) Inappropriate use of ritodrine hydrochloride for threatened preterm birth in Japan: a retrospective cohort study using a national inpatient database. BMC Pregnancy Childbirth 19:204PubMedPubMedCentralCrossRef
19.
Shigemi D, Ishimaru M, Matsui H, Fushimi K, Yasunaga H (2020) Suicide attempts among pregnant and postpartum women in Japan: a nationwide retrospective cohort study. J Clin Psychiatry 81:19m12993PubMedCrossRef
20.
Yasunaga H (2019) Real world data in Japan: Chapter II The Diagnosis Procedure Combination Database. Ann Clin Epidemiol 1:76–9
21.
Shigematsu K, Nakano H, Watanabe Y (2013) The eye response test alone is sufficient to predict stroke outcome—reintroduction of Japan Coma Scale: a cohort study. BMJ Open 3:e002736PubMedPubMedCentralCrossRef
22.
Nakajima M, Okada Y, Sonoo T, Goto T (2023) Development and validation of a novel method for converting the Japan Coma Scale to Glasgow Coma Scale. J Epidemiol 33:531–535PubMedPubMedCentralCrossRef
23.
Makito K, Matsui H, Fushimi K, Yasunaga H (2020) Volatile versus total intravenous anesthesia for cancer prognosis in patients having digestive cancer surgery. Anesthesiology 133:764–773PubMedCrossRef
24.
Mahoney FI, Barthel DW (1965) Functional evaluation: the Barthel index. Md State Med J 14:61–65PubMed
25.
Endo K, Mizuno K, Seki T, Joo WJ, Takeda C, Takeuchi M et al (2022) Intensive care unit versus high-dependency care unit admission on mortality in patients with septic shock: a retrospective cohort study using Japanese claims data. J Intensive Care 10:35PubMedPubMedCentralCrossRef
26.
Doi K, Otani N, Inoue N, Mizuno J, Fushimi K, Yoshino A (2024) Effects of early surgery for cervical fracture dislocation on 30-day mortality using the Japanese diagnosis procedure combination database. Asian Spine J 18:508–513PubMedPubMedCentralCrossRef
27.
Charlson ME, Pompei P, Ales KL, MacKenzie CR (1987) A new method of classifying prognostic comorbidity in longitudinal studies: development and validation. J Chronic Dis 40:373–383PubMedCrossRef
28.
Hurwitz EE, Simon M, Vinta SR, Zehm CF, Shabot SM, Minhajuddin A et al (2017) Adding examples to the ASA-physical status classification improves correct assignment to patients. Anesthesiology 126:614–622PubMedCrossRef
29.
Drake CG (1988) Report of World Federation of Neurological Surgeons Committee on a Universal Subarachnoid Hemorrhage Grading Scale. J Neurosurg 68:985–986
30.
Hunt WE, Hess RM (1968) Surgical risk as related to time of intervention in the repair of intracranial aneurysms. J Neurosurg 28:14–20PubMedCrossRef
31.
Fisher CM, Kistler JP, Davis JM (1980) Relation of cerebral vasospasm to subarachnoid hemorrhage visualized by computerized tomographic scanning. Neurosurgery 6:1–9PubMedCrossRef
32.
Griswold ME, Localio AR, Mulrow C (2010) Propensity score adjustment with multilevel data: setting your sites on decreasing selection bias. Ann Intern Med 152:393–395PubMedCrossRef
33.
Austin PC (2009) Balance diagnostics for comparing the distribution of baseline covariates between treatment groups in propensity-score matched samples. Stat Med 28:3083–3107PubMedPubMedCentralCrossRef
34.
Cameron AC, Gelbach JB, Miller DL (2011) Robust inference with multiway clustering. J Bus Econ Stat 29:238–249CrossRef
35.
Zeileis A, Köll S, Graham N (2020) Various versatile variances: an object-oriented implementation of clustered covariances in R. J Stat Softw 95:1–36CrossRef
36.
Cole SR, Hernán MA (2008) Constructing inverse probability weights for marginal structural models. Am J Epidemiol 168:656–664PubMedPubMedCentralCrossRef
37.
Blum FE, Zuo Z (2013) Volatile anesthetics-induced neuroinflammatory and anti-inflammatory responses. Med Gas Res 3:16PubMedPubMedCentralCrossRef
38.
Athiraman U, Lele AV, Karanikolas M, Dhulipala VB, Jayaraman K, Fong C et al (2022) Inhalational versus intravenous anesthetic conditioning for subarachnoid hemorrhage-induced delayed cerebral ischemia. Stroke 53:904–912PubMedCrossRef
39.
Lauzier DC, Jayaraman K, Yuan JY, Diwan D, Vellimana AK, Osbun JW et al (2023) Early brain injury after subarachnoid hemorrhage: incidence and mechanisms. Stroke 54:1426–1440PubMedPubMedCentralCrossRef
40.
Altay O, Hasegawa Y, Sherchan P, Suzuki H, Khatibi NH, Tang J et al (2012) Isoflurane delays the development of early brain injury after subarachnoid hemorrhage through sphingosine-related pathway activation in mice. Crit Care Med 40:1908–1913PubMedPubMedCentralCrossRef
41.
Voss LJ, Sleigh JW, Barnard JPM, Kirsch HE (2008) The howling cortex: seizures and general anesthetic drugs. Anesth Analg 107:1689–1703PubMedCrossRef
42.
Vychopen M, Lampmann T, Asoglu H, Güresir A, Vatter H, Wach J et al (2023) Non-convulsive status epilepticus in aneurysmal subarachnoid hemorrhage: a prognostic parameter. Brain Sci 13:184PubMedPubMedCentralCrossRef
43.
Lidington D, Wan H, Bolz SS (2021) Cerebral autoregulation in subarachnoid hemorrhage. Front Neurol 12:688362PubMedPubMedCentralCrossRef
44.
Juhász M, Molnár L, Fülesdi B, Végh T, Páll D, Molnár C (2019) Effect of sevoflurane on systemic and cerebral circulation, cerebral autoregulation and CO2 reactivity. BMC Anesthesiol 19:109PubMedPubMedCentralCrossRef
45.
Sorar M, Altay O (2018) Subarachnoid hemorrhage and sevoflurane. Turk Neurosurg 30:354–360
Kritisch Kranke, die einer Intensivbehandlung bedürfen, haben ein hohes Risiko für ernste Beeinträchtigungen nach ihrer Entlassung. Hausarztpraxen kommt hier eine Schlüsselrolle zu.
Behandelnde beurteilen rund 17% der Patientinnen und Patienten in der Primärversorgung als schwierig. Meist handelt es sich um Menschen mit psychischen Begleiterkrankungen, chronischen Schmerzen oder anderen chronischen Leiden. Aber auch die Berufserfahrung der Behandelnden ist für das Urteil relevant.
Beim Abwaschen wurden die Rückenschmerzen jedes Mal unerträglich. Eine 74-Jährige behalf sich mit einem Trick – und erreichte damit, dass ihr nun auch noch der Arm wehtat und die Finger kribbelten.
Als „gefährliches Duo“ bezeichnen zwei Ärzte aus Südkorea die gleichzeitige Anwendung von nichtsteroidalen Antirheumatika (NSAR) und Metformin. Sie schildern den Fall einer älteren Patientin, die mit einer lebensbedrohlichen Laktatazidose in der Notaufnahme vorgestellt wurde.
Mit unserem e.Med Abo Anästhesiologie haben Sie jederzeit Zugriff auf aktuelles, praxisnahes Fachwissen – kompakt und einfach verfügbar. Jetzt CME-Punkte bequem online sammeln.
