Abstract
This is a protocol for a Cochrane Review (Intervention). The objectives are as follows:
To evaluate the effectiveness and safety of thoracic epidural anaesthesia combined with general anaesthesia versus general anaesthesia alone for reducing either the clinical outcomes or shunt fraction consequent from the establishment of OLV.
Background
Description of the condition
One‐lung ventilation (OLV) is the process of excluding one lung from ventilation using airways designed for this purpose. It was developed to improve surgical access for certain thoracic procedures such as lung, oesophageal, aortic or mediastinal surgery (Karzai 2009) and to protect the ventilated lung from contamination with blood or pus from the operated side. OLV was first described in 1931 by Gale and Waters (Brodsky 2005) and quickly resulted in the development of increasingly complex lung resection surgery. OLV is now used for almost all thoracic operations in which the lung is operated on, or in which the collapse of the lung improves access to the operation field. However, during OLV, the adequacy of arterial oxygenation is threatened because blood that flows through the non‐ventilated lung cannot be oxygenated (Bassi 2008). Thus, the blood flow to the non‐ventilated lung is wasted perfusion and blood flowing back to the left heart will be low in oxygen (a right‐to‐left shunt of deoxygenated blood). In addition, there are other factors that may contribute to hypoxia in OLV such as neuromuscular blockade with positive pressure ventilation that favours preferential ventilation of the non‐dependent zones of the lung and inhalational agents that block the hypoxic pulmonary vasoconstriction reflex. Both clinicians and researchers quantify this 'wasted perfusion' as the shunt fraction: the percentage of blood put out by the heart that is not completely oxygenated. There is a normal shunt fraction of about 5% and, theoretically, the right‐to‐left shunt in the non‐ventilated lung plus this normal 5% of blood going through the ventilated lung leads to a shunt fraction in excess of 50% (Lohser 2008) and marked hypoxaemia. On the other hand, both passive and active mechanisms play a role in decreasing the blood flow to the non‐ventilated lung. Surgical manipulation and gravity passively reduce the blood flow to the non‐ventilated lung, and hypoxic pulmonary vasoconstriction actively increases vascular resistance in the non‐ventilated lung, leading to a further decrease in blood flow (Lohser 2008). Multiple reports suggest that hypoxaemia during OLV, defined by a decrease in arterial haemoglobin oxygen saturation (SaO2) to less than 90%, occurs in 5% to 10% of patients (Guenoun 2002; Hurford 1993; Slinger 1988; Slinger 1992). If not appropriately managed in these patients, OLV may result in significant morbidity and mortality from tissue hypoxia.
Description of the intervention
Thoracic epidural anaesthesia is a central neuraxial block technique and involves the use of local anaesthetics injected into the thoracic epidural space to produce a reversible loss of sensation and motor function. It has many benefits, such as reducing the need for intravenous or inhalational analgesia, excellent postoperative analgesia and fast recovery of gastrointestinal function. General anaesthesia refers to the use of a variety of intravenous or inhalational agents (or both) to render the patient unconscious. To date, both thoracic epidural anaesthesia combined with general anaesthesia and general anaesthesia alone have been used for maintaining OLV.
How the intervention might work
Hypoxic pulmonary vasoconstriction (HPV) is an adaptive mechanism unique to the pulmonary circulation that allows redirection of blood flow to alveoli with higher oxygen tension, thereby reducing the blood flow through the non‐ventilated lung (known as a ventilation/perfusion mismatching). HPV is of greatest benefit when 30% to 70% of the lung is made hypoxic (Marshall 1980). During OLV, it is well recognized that volatile anaesthetic agents such as ether, halothane and nitrous oxide inhibit HPV in a dose‐dependent fashion, whereas most intravenous anaesthetics do not inhibit HPV (Benumof 1985; Eisenkraft 1990; Lohser 2008). Thoracic epidural anaesthesia is frequently used for intraoperative and postoperative analgesia in thoracic surgical patients and its role in OLV remains under debate. In experimental studies, epidural anaesthesia did not inhibit HPV (Ishibe 1996; Pfitzner 2006), which may reduce the right‐to‐left shunt in OLV.
Why it is important to do this review
There have been several small randomized controlled trials (RCTs) comparing thoracic epidural anaesthesia combined with general anaesthesia versus general anaesthesia alone for OLV (Garutti 2003; Ozcan 2007; Xu 2010). However, no consistent conclusion has been drawn from these individual studies. We aim to summarize the results of all available data systematically to generate the best evidence for the role of epidural anaesthesia in OLV.
Objectives
To evaluate the effectiveness and safety of thoracic epidural anaesthesia combined with general anaesthesia versus general anaesthesia alone for reducing either the clinical outcomes or shunt fraction consequent from the establishment of OLV.
Methods
Criteria for considering studies for this review
Types of studies
We will include all RCTs that compare thoracic epidural anaesthesia combined with general anaesthesia versus general anaesthesia alone for OLV. We will exclude cross‐over studies and cluster RCTs.
Types of participants
We will include all adult patients (18 years of age or older) undergoing OLV, both emergency and elective cases. We will exclude participants who only have one lung.
Types of interventions
The intervention group will be patients undergoing OLV under thoracic epidural anaesthesia combined with general anaesthesia. The control group will be all patients undergoing OLV under general anaesthesia alone, either maintained with intravenous anaesthetics, inhalational anaesthetics or both.
For inclusion, the trial methodology must include confirmation by direct observation that OLV has been successfully established, and confirmation by fibreoptic bronchoscopy that the tracheal tube is correctly positioned.
Types of outcome measures
Primary outcomes
Hospital mortality.
Major morbidity such as cerebral anoxic injury and myocardial infarction, as defined in the individual included studies.
Secondary outcomes
Length of intensive care unit (ICU) stay: measured in days.
Time to extubation: measured in minutes, defined by discontinuation of propofol infusion/inhaled anaesthesia until extubation.
Hypoxaemia: defined as a decrease in arterial haemoglobin oxygen saturation (SaO2) to less than 90%.
Intrapulmonary shunt fraction. We will calculate shunt fraction according to the formula shown in Appendix 1.
Complications associated with the thoracic epidural anaesthesia such as: cardiac arrest, bleeding, infections, postspinal headache and prolonged neurological (medullary or peripheral) injury lasting more than one month.
Outcomes will not form part of the study eligibility assessment so that studies that meet the participant, intervention and comparison criteria will be included in the review even if they report no relevant outcomes.
Search methods for identification of studies
Electronic searches
We will search the following databases: the Cochrane Central Register of Controlled Trials (CENTRAL) (The Cochrane Library, latest issue), MEDLINE via Ovid SP (1956 to date) and EMBASE via Ovid SP (1980 to present). We will perform electronic searches using exploded Medical Subject Headings and the appropriate corresponding keywords in combination with the sensitivity‐maximizing RCT search strategy (Higgins 2011). We will not apply any language restrictions.
We will search MEDLINE using the search strategy described in Appendix 2. We will adopt our MEDLINE search strategy for searching other databases.
Searching other resources
We will search for ongoing trials in the following registers:
WHO International Clinical Trials Registry Platform;
clinicaltrials.gov;
metaRegister of Controlled Trials;
other national trial registries, such as the Centre for Clinical Trials (China), Clinical Trials Registry (India) and The Netherlands National Trial Register.
We will check the reference lists of all eligible trials and reviews. We will also contact experts in this field in order to identify unpublished research and trials still underway.
Data collection and analysis
Selection of studies
We will combine the results of the searches above and exclude duplicate records. Two authors (FW and YW) will independently screen all titles and abstracts for eligibility. The authors will independently record the reason for excluding each trial. A copy of the 'Study selection form' we will use will be can be found in Appendix 3.
We will resolve any disagreements by consensus or by consulting with a third author (XD) who will arbitrate on trial inclusion. If further information is required to make a decision about trial inclusion, we will contact the first author of the relevant trial (LT).
We will compile a list of all eligible trials. (See Appendix 4 for a copy of the 'Form for eligible trials').
Data extraction and management
Two authors (FW and YW) will independently extract data using a standardized data extraction form. A copy of the form we will use can be found in Appendix 5. We will resolve any disagreements by discussion and, if necessary, by consulting with a third author (XD). If further information is required, we will contact corresponding authors via email (LT).
Assessment of risk of bias in included studies
Two authors (FW and YW) will independently assess the methodological quality of the eligible trials using the 'Risk of bias' tool described in Chapter 8 of the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2011). A copy of the form we will use can be found in Appendix 6. We will resolve disagreements by discussion, where necessary, with the help of a third review author (XD).
We will use the following criteria to assess the risk of bias of the included studies: the quality domains of random sequence generation, allocation concealment, blinding of participants and personnel, blinding of outcome assessment, incomplete outcome data, selective reporting and any other potential threats to validity.
We will consider a trial as having a low risk of bias if all domains are assessed as adequate. We will consider a trial as having a high risk of bias if one or more domain is assessed as inadequate or unclear. We will report the 'Risk of bias' table as part of the 'Characteristics of included studies' table and present a 'Risk of bias' summary figure which will detail all of the judgements made for all included studies in the review.
Measures of treatment effect
We will use odds ratio (OR) as a measure of effect for dichotomous outcomes (such as the incidence of hypoxaemia). We will use mean difference (MD) for continuous variables.
Unit of analysis issues
The individual participant in each clinical trial will be the unit of analysis. We will exclude cross‐over studies and cluster RCTs.
Dealing with missing data
We will contact the authors of included trials to obtain missing data necessary for meta‐analysis (LT). If this fails, we will calculate standard deviations from the standard errors or confidence intervals, as described in the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2011). Where standard deviations cannot be calculated, we will impute these using the mean of the reported standard deviations from the other trials.
We will address the impact of missing data in the discussion section of the review.
Assessment of heterogeneity
We will assess the clinical heterogeneity of the included studies according to their clinical and methodological diversity ('Risk of bias' assessment). We will address clinical heterogeneity using subgroup analyses.
We will assess statistical heterogeneity using visual inspection of the forest plot and the I2 statistic. We will consider significant statistical heterogeneity to be present when the I2 value is greater than 50%.
Assessment of reporting biases
We will construct funnel plots to assess any publication bias if more than 10 trials are included in our systematic review.
Data synthesis
If the degree of heterogeneity is not excessive, we will generate a quantitative summary by meta‐analysis using Review Manager software (RevMan 5.1). We will use a fixed‐effect model if there is no significant heterogeneity among studies. We will use a random‐effects model if the P value for the Cochran Q statistic is less than 0.1.
Subgroup analysis and investigation of heterogeneity
We will perform subgroup analysis for different measured time points and agents of general anaesthesia and epidural anaesthesia. We will compare subgroups using an interaction term if appropriate.
Subgroup analysis according to measured time point:
10 minutes after one‐lung ventilation
15 minutes after one‐lung ventilation
20 minutes after one‐lung ventilation
30 minutes after one‐lung ventilation
Subgroup analysis according to agents of general anaesthesia:
Control patients received intravenous general anaesthesia
Control patients received inhalational general anaesthesia
Control patients received both intravenous and inhalational general anaesthesia
Subgroup analysis according to agents of epidural anaesthesia:
Patients received opiates alone for epidural anaesthesia
Patients received local anaesthetics alone for epidural anaesthesia
Patients received mixed agents for epidural anaesthesia
Sensitivity analysis
We will perform sensitivity analysis to exclude trials at high risk of bias or trials with significant clinical heterogeneity. We will compare random‐effects and fixed‐effect estimates of each outcome variable. To assess the influence of the trials which have imputed standard deviation (SD) data, we will perform sensitivity analyses based on the 'borrowed' data. If sensitivity analyses identify particular factors that influence the conclusions of the review, we will explore the potential causes of the uncertainties and interpret the result of the review with caution.
Summary of findings
We will use the principles of the GRADE system (Guyatt 2008) to assess the quality of the body of evidence associated with the specific outcomes intrapulmonary shunting, hypoxaemia, time to extubation, hospital mortality, length of ICU stay and complications associated with thoracic epidural anaesthesia in our review and construct a 'Summary of findings' (SOF) table using the GRADE software. The GRADE approach appraises the quality of a body of evidence based on the extent to which one can be confident that an estimate of effect or association reflects the item being assessed. The quality of a body of evidence considers the study methodological quality, the directness of the evidence, heterogeneity of the data, precision of the effect estimates and the risks of publication bias.
Acknowledgements
We would like to thank Mike Bennett (content editor), Marialena Trivella (statistical editor), Javier Eslava‐Schmalbach and Joanne Guay (peer reviewers) for their help and editorial advice during the preparation of this protocol for the systematic review.
We also want to thank Jane Cracknell (Manager Editor, Cochrane Anaesthesia Review Group (CARG)) for her assistance in the preparation of this protocol.
Appendices
Appendix 1. Shunt fraction
The shunt fraction (Qs/Qt) will be calculated from the formula:
Qs/Qt = (CcO2 ‐ CaO2)/(CcO2 ‐ CvO2)
CcO2 = (Hb * 1.34) + (PAO2 * 0.0031)
C (a or v) O2 = (Hb * 1.34 * %Saturation) + (0.0031 * PO2)
PAO2 = FiO2 (Pb ‐ PH2O) ‐ (PaCO2/0.8)
Qs means the amount of shunted blood (blood not participating in gas exchange).
Qt means the total cardiac output.
CcO2 means the oxygen content of pulmonary capillary blood.
CaO2 means oxygen content of arterial blood.
CvO2 means oxygen content of mixed venous blood.
PAO2 means partial pressure of oxygen in alveolar air.
PaO2 means partial pressure of oxygen in arterial blood.
PvO2 means partial pressure of oxygen in mixed venous blood.
Hb = haemoglobin.
%Saturation = % oxygen saturation in arterial blood or mixed venous blood.
FiO2 = fraction of inspiration O2.
Pb = pressure of atmosphere.
PaCO2 = partial pressure of carbon dioxide in arterial blood.
Appendix 2. OVID MEDLINE search strategy
1 exp Respiration, Artificial/ or exp One‐Lung Ventilation/ or ((one or single) adj3 lung*).af. 2 ((thoracic or epidural) adj3 an?esth*).ti,ab. or exp Anesthesia, Epidural/ or exp Anesthesia, General/ or exp Anesthesia, Inhalation/ or exp Anesthesia, Intravenous/ 3 1 and 2 4 ((randomized controlled trial or controlled clinical trial).pt. or randomized.ab. or placebo.ab. or clinical trials as topic.sh. or randomly.ab. or trial.ti.) not (animals not (humans and animals)).sh. 5 3 and 4
Appendix 3. Study selection form
Thoracic epidural anaesthesia combined with general anaesthesia versus general anaesthesia alone for one‐lung ventilation
Study selection form
1. General information
| First author | Year | Publication type | Reviewer | Review date | Title reviewed | Abstract reviewed | Full text reviewed |
2. Study selection guidelines
| Study characteristics | Eligibility criteria | Result | ||
| Yes | No | Unclear | ||
| Types of studies | RCT | |||
| Types of participants | Adult patients (18 years of age or older) undergoing an elective one‐lung ventilation surgical procedure | |||
| Types of intervention | The control group will be all patients undergoing general anaesthesia, either intravenous anaesthesia or inhalational anaesthesia The intervention group will be all patients undergoing thoracic epidural anaesthesia combined with general anaesthesia |
|||
| Types of outcome measures | Relevant outcomes (intrapulmonary shunting, hospital mortality, time to extubation, length of ICU stay, hypoxaemia, complications associated with thoracic epidural anaesthesia) | |||
| Included Excluded | ||||
| Reason for exclusion | ||||
Appendix 4. Eligible trials form
Thoracic epidural anaesthesia combined with general anaesthesia versus general anaesthesia alone for one‐lung ventilation
Eligible trials form
| Unique ID | ISDN | Author(s) | Publication type | Year |
| 1 | ||||
| 2 | ||||
| 3 |
Appendix 5. Data extraction form
Thoracic epidural anaesthesia combined with general anaesthesia versus general anaesthesia alone for one‐lung ventilation
Data extraction form
| Unique ID | ISDN | First author | Year | Reviewer | Review date |
Study methods and details
| Trial characteristics | Details |
| Country/countries | |
| Participant inclusion criteria | |
| Participant exclusion criteria | |
| Mean or median age of participants | |
| Gender of participants | |
| Baseline imbalances | |
| Surgery information | |
| How many participants were randomized? | |
| Number of participants allocated to: Combined anaesthesia General anaesthesia alone |
|
| Number of participants analysed: Combined anaesthesia General anaesthesia alone |
|
| Combined anaesthesia method | |
| General anaesthesia alone method | |
| Outcome definition |
Outcomes
For continuous data
| Outcome | Combined anaesthesia group | General anaesthesia alone group | ||||
| n | Mean | SD | n | Mean | SD | |
| Qs/Qt (10 minutes) | ||||||
| Qs/Qt (15 minutes) | ||||||
| Qs/Qt (20 minutes) | ||||||
| Qs/Qt (30 minutes) | ||||||
| PaO2 (10 minutes) | ||||||
| PaO2 (15 minutes) | ||||||
| PaO2 (20 minutes) | ||||||
| PaO2 (30 minutes) | ||||||
| Length of ICU stay | ||||||
| Time to extubation | ||||||
| Length of hospital stay | ||||||
| Vasopressin requirement | ||||||
For dichotomous data
| Outcome | Combined anaesthesia | General anaesthesia alone | ||
| Number with event | Number without event | Number with event | Number without event | |
| Hypoxia | ||||
| Bleeding | ||||
| Infections | ||||
| Postspinal headache | ||||
| Pneumonia | ||||
| Hospital mortality | ||||
Appendix 6. Quality assessment form
Thoracic epidural anaesthesia combined with general anaesthesia versus general anaesthesia alone for one‐lung ventilation
Quality assessment form
| Unique ID | ISDN | First author | Year | Reviewer | Review date |
| Domain | Describe | Risk of bias |
| 1. Adequate sequence generation | Low/High/Unclear | |
| 2. Allocation concealment | Low/High/Unclear | |
| 3. Blinding of participants and personnel | Low/High/Unclear | |
| Blinding of outcome assessment | Low/High/Unclear | |
| 4. Incomplete outcome data | Low/High/Unclear | |
| 5. Selective outcome reporting | Low/High/Unclear | |
| 6. Other potential threats to validity | Low/High/Unclear |
Contributions of authors
Fei Wang (FW), Youping Wu (YW), Lu Tang (LT), Chengjie Gao (CG), Jinbao Li (JL) and Xiaoming Deng (XD)
FW, YW and LT contribute equally and they are all first authors. JL and XD are co‐corresponding authors.
Conceiving the review: FW, YW, LT, CG, JL, XD
Co‐ordinating the review: JL, XD
Undertaking manual searches: FW, YW
Screening search results: FW, YW
Organizing retrieval of papers: LT
Screening retrieved papers against inclusion criteria: FW, YW
Appraising quality of papers: FW, YW
Abstracting data from papers: FW, YW
Writing to authors of papers for additional information: FW, YW
Providing additional data about papers: FW, YW
Obtaining and screening data on unpublished studies: FW, YW
Data management for the review: LT, CG
Entering data into Review Manager (RevMan 5.1): FW, YW, LT, CG
RevMan statistical data: FW, YW, LT, CG
Other statistical analysis not using RevMan: FW, YW, LT, CG
Interpretation of data: FW, YW
Statistical inferences: FW, YW
Writing the review: FW, YW, LT, JL, CG, XD
Securing funding for the review: JL, XD
Performing previous work that was the foundation of the present study: FW, YW, LT, CG, JL, XD
Guarantor for the review (one author): FW
Person responsible for reading and checking the review before submission: FW
Declarations of interest
Fei Wang: none known
Youping Wu: none known
Lu Tang: none known
Chengjie Gao: none known
Jinbao Li: none known
Xiaoming Deng: none known
Notes
February 2017
The protocol is being withdrawn because it is out of date and does not meet the current methodological standards of The Cochrane Collaboration
Withdrawn from publication for reasons stated in the review
References
Additional references
- Bassi A, Milani WRO, Dib RP, Matos D. Intravenous versus inhalation anaesthesia for one‐lung ventilation. Cochrane Database of Systematic Reviews 2008, Issue 2. [DOI: 10.1002/14651858.CD006313.pub2] [DOI] [PubMed] [Google Scholar]
- Benumof JL. One‐lung ventilation and hypoxic pulmonary vasoconstriction: implications for anesthetic management. Anesthesia and Analgesia 1985;64(8):821‐33. [PUBMED: 3893223] [PubMed] [Google Scholar]
- Brodsky JB. The evolution of thoracic anesthesia. Thoracic Surgery Clinics 2005;15(1):1‐10. [PUBMED: 15707342] [DOI] [PubMed] [Google Scholar]
- Eisenkraft JB. Effects of anaesthetics on the pulmonary circulation. British Journal of Anaesthesia 1990;65(1):63‐78. [PUBMED: 2200486] [DOI] [PubMed] [Google Scholar]
- Garutti I, Cruz P, Olmedilla L, Barrio JM, Cruz A, Fernandez C, et al. Effects of thoracic epidural meperidine on arterial oxygenation during one‐lung ventilation in thoracic surgery. Journal of Cardiothoracic and Vascular Anesthesia 2003;17(3):302‐5. [PUBMED: 12827575] [DOI] [PubMed] [Google Scholar]
- Guenoun T, Journois D, Silleran‐Chassany J, Frappier J, D'attellis N, Salem A, et al. Prediction of arterial oxygen tension during one‐lung ventilation: analysis of preoperative and intraoperative variables. Journal of Cardiothoracic and Vascular Anesthesia 2002;16(2):199‐203. [PUBMED: 11957171] [DOI] [PubMed] [Google Scholar]
- Guyatt GH, Oxman AD, Kunz R, Vist GE, Falck‐Ytter Y, Schunemann HJ. What is "quality of evidence" and why is it important to clinicians?. BMJ (Clinical research ed.) 2008;336(7651):995‐8. [PUBMED: 18456631] [DOI] [PMC free article] [PubMed] [Google Scholar]
- Higgins JPT, Green S (editors). Cochrane Handbook for Systematic Reviews of Interventions Version 5.1.0 [updated March 2011]. The Cochrane Collaboration, 2011. Available from www.cochrane‐handbook.org.
- Hurford WE, Alfille PH. A quality improvement study of the placement and complications of double‐lumen endobronchial tubes. Journal of Cardiothoracic and Vascular Anesthesia 1993;7(5):517‐20. [PUBMED: 8268428] [DOI] [PubMed] [Google Scholar]
- Ishibe Y, Shiokawa Y, Umeda T, Uno H, Nakamura M, Izumi T. The effect of thoracic epidural anesthesia on hypoxic pulmonary vasoconstriction in dogs: an analysis of the pressure‐flow curve. Anesthesia and Analgesia 1996;82(5):1049‐55. [PUBMED: 8610866] [DOI] [PubMed] [Google Scholar]
- Karzai W, Schwarzkopf K. Hypoxemia during one‐lung ventilation: prediction, prevention, and treatment. Anesthesiology 2009;110(6):1402‐11. [PUBMED: 19417615] [DOI] [PubMed] [Google Scholar]
- Lohser J. Evidence‐based management of one‐lung ventilation. Anesthesiology Clinics 2008;26(2):241‐72, v. [PUBMED: 18456211] [DOI] [PubMed] [Google Scholar]
- Marshall BE, Marshall C. Continuity of response to hypoxic pulmonary vasoconstriction. Journal of Applied Physiology: Respiratory, Environmental and Exercise Physiology 1980;49:189‐96. [PUBMED: 7400002] [DOI] [PubMed] [Google Scholar]
- Ozcan PE, Senturk M, Sungur Ulke Z, Toker A, Dilege S, Ozden E, et al. Effects of thoracic epidural anaesthesia on pulmonary venous admixture and oxygenation during one‐lung ventilation. Acta Anaesthesiologica Scandinavica 2007;51(8):1117‐22. [PUBMED: 17697309] [DOI] [PubMed] [Google Scholar]
- Pfitzner J. Potential for acute lung injury following one‐lung ventilation: alveolar overdistension from partial bronchial obstruction. Anaesthesia2006; Vol. 61, issue 9:906‐7. [PUBMED: 16922766] [DOI] [PubMed]
- The Nordic Cochrane Centre, The Cochrane Collaboration. Review Manager (RevMan). Version 5.1. Copenhagen: The Nordic Cochrane Centre, The Cochrane Collaboration, 2011.
- Slinger P, Triolet W, Wilson J. Improving arterial oxygenation during one‐lung ventilation. Anesthesiology 1988;68(2):291‐5. [PUBMED: 3277487] [DOI] [PubMed] [Google Scholar]
- Slinger P, Suissa S, Triolet W. Predicting arterial oxygenation during one‐lung anaesthesia. Canadian Journal of Anaesthesia 1992;39(10):1030‐5. [PUBMED: 1464128] [DOI] [PubMed] [Google Scholar]
- Xu Y, Tan Z, Wang S, Shao H, Zhu X. Effect of thoracic epidural anesthesia with different concentrations of ropivacaine on arterial oxygenation during one‐lung ventilation. Anesthesiology 2010;112(5):1146‐54. [PUBMED: 20418695] [DOI] [PubMed] [Google Scholar]