Glutathione oxidation correlates with one-lung ventilation time and PO₂/FiO₂ ratio during pulmonary lobectomy

Abstract

Objectives: During lung lobectomy, the operated lung completely collapses with simultaneous hypoxic pulmonary vasoconstriction, followed by expansion and reperfusion. Here, we investigated glutathione oxidation and lipoperoxidation in patients undergoing lung lobectomy, during one-lung ventilation (OLV) and after resuming two-lung ventilation (TLV), and examined the relationship with OLV duration.

Methods: We performed a single-centre, observational, prospective study in 32 patients undergoing lung lobectomy. Blood samples were collected at five time-points: T0, pre-operatively; T1, during OLV, 5 minutes before resuming TLV; and T2, T3, and T4, respectively, 5, 60, and 180 minutes after resuming TLV. Samples were tested for reduced glutathione (GSH), oxidized glutathione (GSSG), glutathione redox potential, and malondialdehyde (MDA).

Results: GSSG and MDA blood levels increased at T1, and increased further at T2. OLV duration directly correlated with marker levels at T1 and T2. Blood levels of GSH and glutathione redox potential decreased at T1−T3. GSSG, oxidized glutathione/total glutathione ratio, and MDA levels were inversely correlated with arterial blood PO₂/FiO₂ at T1 and T2.

Discussion: During lung lobectomy and OLV, glutathione oxidation, and lipoperoxidation marker blood levels increase, with further increases after resuming TLV. Oxidative stress degree was directly correlated with OLV duration, and inversely correlated with arterial blood PO₂/FiO₂.

Keywords:

• Lung lobectomy
• One-lung ventilation
• Oxidative lung injury
• Glutathione oxidation
• Glutathione redox potential
• Malondialdehyde

View correction statement:

Erratum

Introduction

During pulmonary lobectomy, one-lung ventilation (OLV) is applied because the operated lung is completely collapsed, hypoperfused, and suffers reactive hypoxic pulmonary vasoconstriction in response to alveolar hypoxia.^1,2 Moreover, the hypoxic tissues increase reactive oxygen species (ROS) production in the mitochondria respiratory chain,^3,4 contributing to increasing vasoconstriction.⁵ Altogether, patients undergoing lobar lung resection suffer relative ischaemia during lung collapse, followed by expansion-reperfusion and oxidative injury attributed to ROS.^6,7 However, few studies have examined this sequence of events. Increased urinary malondialdehyde (MDA) levels are reported in patients following pulmonary resection, and studies show increased blood MDA levels after lung resection and resuming two-lung ventilation (TLV), with massive superoxide production associated with lung re-expansion and reoxygenation.^8–11 To date, no studies have investigated the time-course of blood glutathione (GSH) oxidation in patients undergoing lung lobectomy or its relationship with lung collapse duration.

 GSH is the predominant low-molecular-weight thiol in animal cells. It originates primarily from the liver and is mainly detected in the cytosol at concentrations ranging from 0.1 to 10 mM, with the highest levels reported in the liver, spleen, kidney, lens, erythrocytes, and leukocytes.¹² Plasma concentrations of reduced GSH are relatively low (2–20 μmol/l) because the cysteine residue is readily oxidized by free radicals and reactive oxygen/nitrogen species to form glutathione disulphide (GSSG). GSH is involved in antioxidant defence and several other cellular functions. The GSSG/2GSH system is the most important cellular redox coupling, playing crucial roles in antioxidant protection.¹² Blood glutathione levels may reflect the glutathione status throughout the body, including in other less accessible tissues. Thus, measurements of GSH and GSSG in whole blood are considered an index of oxidative injury and a potential indication of disease.^121314–15

Lipid peroxidation in cell membranes can be defined as the oxidative deterioration of lipids containing carbon–carbon double bonds. A large number of metabolites are formed, including many electrophiles that react with protein and DNA. The aldehyde MDA is the principal and most studied product of polyunsaturated fatty acid peroxidation. MDA is a highly toxic molecule that contributes to reperfusion damage.^9,10,16,17

The present study aimed to investigate the time-course of glutathione oxidation and lipid peroxidation in blood from patients undergoing lung lobectomy, during OLV and after resuming TLV. We also examined the correlation between OLV duration and oxidative stress marker levels. Finally, we studied the relationship between glutathione oxidation and arterial blood PO₂/FiO₂ values immediately before and after resuming TLV, as a marker of poor oxygenation of the operated lung and hypoxaemia before and after OLV.

Material and methods

Study subjects

This prospective observational study included 32 lung cancer patients (27 men and 5 women) with non-small cell lung cancer who were scheduled for elective lung lobectomy at a single centre. Exclusion criteria included previous radiotherapy or chemotherapy, oral antioxidant vitamin use, active systemic infection, body temperature above 38°C, pregnancy, drug addiction, malnutrition, or body weight loss of >20% in the last 2 months. Each patient gave informed consent, and the study protocol was approved by the Ethics Committee at Hospital Clínico Universitario de Valencia.

Anaesthetics details

Anaesthesia was induced intravenously using sodium thiopental (3–5 mg/kg) and fentanyl (2 μg/kg). Tracheal intubation was performed with a double-lumen tube and using rocuronium bromide (0.6 mg/kg). Anaesthesia was maintained with sevoflurane inhalation. During anaesthesia, we monitored arterial oxygen saturation, invasive blood pressure in the radial artery, heart rate, and end-tidal CO₂.

TLV was performed in a volume-controlled mode with a tidal volume of 7–8 ml/kg ideal body weight, using a respiratory rate of 12–14 breaths/min a 1:2 I:E ratio. Once the operated lung was collapsed and the thorax was opened to air, OLV was performed with a tidal volume of 5–6 ml/kg ideal body weight, a 14–18 breaths/min respiratory rate, and a 1:2 I:E ratio, maintaining an end-tidal CO₂ of <40 mmHg. To keep O₂ saturation >92% during TLV and OLV, we used a positive end-expiratory pressure (PEEP) of 5–7 cm H₂O and a 50% fraction of inspired oxygen (FiO₂ 0.5). FiO₂ was increased momentarily if O₂ saturation decreased to <92% and not efficacy of recruitment manoeuvres to the ventilated lung. For post-operative pain management, a thoracic epidural catheter was inserted. After surgery, all patients were transferred to the care unit for at least 24 hours.

Blood sampling

From each patient, five arterial blood samples were collected from a radial artery line at specific time-points: T0, pre-operatively; T1, during OLV, 5 minutes before resuming TLV; and T2, T3, and T4, respectively, 5, 60, and 180 minutes after resuming TLV. An ABL 88 Flex apparatus (Radiometer, Denmark) was used to analyse the following arterial blood gas parameters: pH, partial oxygen pressure (PaO₂), PaO₂/FiO₂ ratio, and partial carbon dioxide pressure (PaCO₂).

Determination of oxidative stress markers levels in blood

To analyse GSSG in the presence of a large excess of GSH, proteins were precipitated from arterial whole blood samples (0.5 ml) using 0.5 ml ice-cold perchloric acid (12%) containing 20 mM of the GSH quenching agent N-ethylmaleimide (Sigma Chemical Co., St Louis, MO, USA) to prevent GSH oxidation during sample preparation, and 1 mM of the metal chelator bathophenanthrolinedisulfonic acid (Sigma Chemical Co.). Samples were then centrifuged at 15 000 g for 5 minutes at 4°C. The acidic supernatants were stored frozen at −20°C until derivatization, and were then used to determine GSSG levels using high-performance liquid chromatography (HPLC) with UV-visible detection (365 nm).¹⁸

To analyse GSH, arterial whole blood samples (0.5 ml) were added to 0.5 ml ice-cold 30% trichloroacetic acid containing 2 mM ethylenediaminetetraacetic acid. Samples were then centrifuged at 15 000 g for 5 minutes at 4°C, and the acidic supernatant was stored at −20°C until analysis. GSH was measured using the glutathione-S-transferase (GST) assay, based on the GST-catalysed reaction between chlorodinitrobenzene and GSH in 0.1 M potassium phosphate buffer (pH 7), which was spectrophotometrically detected at 340 nm.¹⁹ The oxidized glutathione/total glutathione ratio was calculated as [2GSSG/(GSH + 2GSSG)]*100 and expressed as a percentage that indicates the glutathione redox status and is an oxidative stress index.

To analyse MDA, arterial whole blood samples (1 ml) were collected from a radial artery line into BD Vacutainer LH PST II tubes with lithium heparin (Plymouth, UK). Samples were centrifuged at 3000g for 12 minutes, and the separated plasma was stored at −80°C until analysis. Lipid peroxides were hydrolysed by boiling in diluted phosphoric acid. Then MDA was reacted with thiobarbituric acid to form the MDA(TBA)₂ adduct, which was detected by HPLC and quantified with UV light at 532 nm.²⁰

Calculation of redox potential

To calculate the redox state of the GSSG–GSH couple, we used the Nernst equation: E_h = E₀ + RT/nF ln[disulphide]/[thiol]², where E_o is the standard potential for the redox couple, R is the gas constant, T is the absolute temperature, n is 2 for the number of electrons transferred, and F is Faraday’ s constant. To calculate the redox potential in whole blood, we used the equation E_h (mV) = E₀ + 30 log ([GSSG]/[GSH]²) at 37°C (310 K), where the standard potential E₀ is −264 mV at pH 7.4 (−5.9 mV for every 0.1 increase in pH), E₀ at pH 7.0 is −240, and the concentrations are in moles/liter.^21,22

Statistical analysis

Results are expressed as mean ± SD. Data were parametric and normally distributed. One-way analysis of variance for repeated measures (T1–T4 vs. T0) was performed to assess the effects of OLV time and subsequent lung re-expansion on blood levels of glutathione oxidation and lipid peroxidation markers. Then, the data sets in which F was significant were examined by the t-test, using P < 0.05 as the critical limit. Pearson's correlation coefficient (r) was calculated to analyse how total lung collapse duration correlated with blood levels of oxidative stress markers and redox potential or with the PaO₂/FiO₂ ratio. Analyses were performed using SPSS version 11.0 (Chicago, IL, USA). Excel (Microsoft Office 2011, Redmond, WA, USA) was used for graphics.

Results

Table 1 summarizes pre-operative parameters and clinical data during surgery. Seven patients intermittently required >50% oxygen during OLV to maintain O₂ saturation of >92%. No patients had difficulty maintaining lung isolation during surgery. In five patients, the PaO₂/FiO₂ ratio in arterial blood gas was <200 between the first 24 hours of admission to in the post-operative care unit (mean for all patients, 250 ± 54).

Table 1 Demographic and clinical data (n = 32)

Age (years)	65.6 ± 10.0
Sex (M/F), n	27/5
BMI (kg/m^2)	25.8 ± 4.3
Current smokers, n	10
Ex-smokers, n	19
Never smoked, n	3
Pre-operative FEV1 (l)	2.05 ± 0.53
Side (right/left), n	20/12
Histology, n
Adenocarcinoma	13
Squamous carcinoma	18
Other	1
OLV time (min)	111 ± 38
Surgery duration (min)	150 ± 37
Intraoperative fluid load (l)	1.28 ± 0.30
Intraoperative urine (l)	0.60 ± 0.38
Average mean arterial pressure (mmHg)	78 ± 9
Average heart rate (beats/min)	84 ± 16
Average temperature (°C)	35.7 ± 0.6

Values expressed as mean ± SD unless otherwise noted. BMI, body mass index; FEV₁, forced expiratory volume in 1second; OLV, one-lung ventilation.

Table 2 summarizes the arterial blood gas parameters at five time-points. Comparing T1–T4 values to T0 values revealed a significant decrease of pH at T2 (5 minutes after resuming TLV; P = 0.041). At T1 (during OLV, 5 minutes before resuming TLV), we observed a significant increase in PaCO₂ (P < 0.001). PaO₂ values decreased at T1 (P<0.001), T2 (P < 0.01), T3 (P < 0.01), and T4 (P < 0.001). Thus, PaO₂/FiO₂ also significantly decreased at T1 (P < 0.001), T2 (P < 0.01), T3 (P < 0.05), and T4 (P < 0.01).

Table 2 Arterial blood gas parameters during lung lobectomy (n = 32)

Time-points	T0	T1	T2	T3	T4
pH	7.373 ± 0.034	7.359 ± 0.026	7.356 ± 0.033*	7.373 ± 0.027	7.368 ± 0.042
PaCO2	36.5 ± 3.2	39.2 ± 2.3***	37.7 ± 2.7	37.3 ± 1.6	36.9 ± 3.3
PaO2	171 ± 30	134 ± 30***	147 ± 33**	145 ± 31**	146 ± 25***
PaO2/FiO2	333 ± 68	243 ± 76***	278 ± 71**	290 ± 62*	291 ± 50**

Values expressed as mean ± SD. Differences T1–T4 vs. T0 were analysed with two-tailed Student's t-test.

PaO₂ (mmHg), oxygen partial pressure; PaCO₂ (mmHg), carbon dioxide partial pressure; FiO₂, fraction of inspired oxygen.

*P < 0.05.

**P < 0.01.

***P < 0.001.

Figure 1 summarizes the time course of oxidative stress marker levels in arterial blood from T0 to T4. The main cellular antioxidant GSH decreased significantly during time-points T1 (P < 0.01), T2 (P < 0.001), and T3 (P < 0.01) (Fig. 1(A)). In contrast, blood GSSG levels significantly increased at T1 (P < 0.001), T2 (P < 0.001), T3 (P < 0.001), and T4 (P < 0.01) (Fig. 1(B)), with a particular increase at T2 due to strong GSH oxidation during lung re-expansion. The blood oxidized glutathione/total glutathione ratio was also increased at T1 (P < 0.001), T2 (P < 0.001), and T3 (P < 0.001) (Fig. 1(C)), which represents the systemic redox status. MDA, an end-product of cellular phospholipids peroxidation, also increased at T1 (P < 0.01), T2 (P < 0.001), T3 (P < 0.01), and T4 (P < 0.05) – most intensely at T2 due to lung re-expansion (Fig. 1(D)).

Figure 1 Time-course of glutathione oxidation and lipid peroxidation marker levels in arterial blood. (A) Reduced glutathione (GSH). (B) Oxidized glutathione (GSSG). (C) Oxidized glutathione/total glutathione ratio ([2GSSG/(GSH + 2GSSG)]*100). (D) Malondialdehyde (MDA). T0, pre-operatively; T1, 5 minutes before resuming two-lung ventilation (TLV);T2, 5 minutes after resuming TLV;T3, 60 minutes after resuming TLV;T4, 180 minutes after resuming TLV. Values are expressed as mean ± SD (n = 32). Differences between T0 and T1–T4 were analysed using a two-tailed Student's t-test. *P < 0.05;**P < 0.01;***P < 0.001.

Table 3 presents the time-course of redox potentials for the GSSH–GSH couple in whole blood. Compared to at T0, the redox potentials at T1–T4 showed significantly increased oxidation. Redox potential was most oxidized at T2, after lung re-expansion immediately following the lobectomy, in parallel with GSH and GSSG levels (Fig. 1). Furthermore, we found a significant inverse correlation of lung collapse duration with redox potential at T1–T3, particularly at T2 (r = −0.4888; P = 0.0039). Redox potentials were also correlated with arterial blood PO₂/FiO₂ values, particularly at T1 (r = 0.3627; P = 0.0380) and T4 (r = 0.4034; P = 0.0199) (Table 3).

Table 3 Redox potential of the GSSG/2GSH couple, and the correlations with lung collapse time and PaO₂/FiO₂ ratio during lung lobectomy

Time-points	T0	T1	T2	T3	T4
Eh (mV)	−209 ± 7.6	−199 ± 7.1	−197 ± 8.0	−200 ± 8.3	−203 ± 8.0
		P < 0.0001	P < 0.0001	P < 0.0001	P = 0.0018
rmin	−0.0145	−0.3504	−0.4888	−0.3811	−0.0538
	P = 0.9362	P = 0.0456	P = 0.0039	P = 0.0287	P = 0.7662
rP/F	−0.0273	0.3627	0.2989	0.0094	0.4034
	P = 0.8803	P = 0.0380	P = 0.0911	P = 0.9586	P = 0.0199

Values expressed as mean ± SD (n = 32). E_h, redox potential of GSSG/2GSH couple; r_min, Pearson's correlation coefficient between E_h and lung collapse time; r_P/F, Pearson's correlation coefficient between E_h and PaO₂/FiO₂ values; PaO₂ (mmHg): oxygen partial pressure; FiO₂: fraction of inspired oxygen. For E_h, differences of T1–T4 vs. T0 were analysed with a two-tailed Student's t-test. P values of <0.05 were considered statistically significant.

Total lung collapse duration was positively correlated with blood oxidized glutathione levels at T1 (r = 0.611; P = 0.0002), and this correlation was stronger at T2 (r = 0.652; P = 0.0001) (Fig. 2(A)). However, reduced glutathione levels were not correlated with OLV times (Fig. 2(B)). Oxidized glutathione/total glutathione ratio was positively correlated with OLV times at T1 (r = 0.383; P = 0.0306) and T2 (r = 0.436; P = 0.0129) (Fig. 2(C)). Lung collapse duration was also strongly significantly correlated with MDA levels at T1 (r = 0.583; P = 0.0005) and T2 (r = 0.738; P = 0.0001) (Fig. 2(D)).

Figure 2 Pearson's correlation coefficients (r) were determined between the total duration of lung collapse (min) and the oxidative stress marker levels in the blood at time-points T1 (5 minutes before resuming two-lung ventilation) and T2 (5 minutes after resuming two-lung ventilation) (n = 32). (A) Oxidized glutathione (GSSG). (B) Reduced glutathione (GSH). (C) Oxidized glutathione/total glutathione ratio ([2GSSG/(GSH + 2GSSG)]*100). (D) Malondialdehyde (MDA). P-values of <0.05 were considered statistically significant.

Furthermore, we found that blood GSSG, oxidized glutathione/total glutathione ratio, and MDA levels were significantly inversely correlation with arterial blood PO₂/FiO₂ values at T1 and T2 (Fig. 3(A–C)). Lung collapse duration was also inversely correlated with arterial blood PO₂/FiO₂ values, particularly at T1 (Fig. 3(D)).

Figure 3 Pearson's correlation coefficients (r) were determined between PO₂/FiO₂ values in arterial blood and total time of lung collapse (min), and between PO₂/FiO₂ values in arterial blood and glutathione oxidation and lipid peroxidation marker levels in blood at T1 (5 minutes before resuming two-lung ventilation) and T2 (5 minutes after resuming two-lung ventilation). (A) Oxidized glutathione (GSSG). (B) Oxidized glutathione/total glutathione ratio ([2GSSG/(GSH + 2GSSG)]*100). (C) Malondialdehyde (MDA). (D) Total time of lung collapse (min). P-values of <0.05 were considered statistically significant.

Discussion

The pathological mechanisms of lung injury and oedema following pulmonary resection are considered to be related to the effects of ROS in the operated lung parenchyma, at least partly due to collapse, with hypoxic insult followed by re-oxygenation injury after lung re-expansion.^{6–11,2324–25} The operated lung is totally atelectatic with only a small fraction of pulmonary blood flow passing through it. After resuming TLV, the operated lung may suffer reduced perfusion due to microatelectasis-induced vasoconstriction.²⁶

Examining arterial blood gas parameters at time-points throughout lung lobectomy revealed a reperfusion-associated pH decrease at 5 minutes after resuming TLV (T2). Moreover, PaCO₂ moderately increased due to hypoventilation at T1, 5 minutes before resuming TLV. Relative to T0, PaO₂ decreased at all time-points, particularly in T1, during operated lung collapse. PaO₂/FiO₂ also decreased at T1–T4. These findings suggest a ventilation/perfusion mismatch during and after operated lung collapse.

Blood levels of GSH – the main molecule involved in antioxidant defence, ROS scavenging, and removing hydrogen and lipid peroxides – were decreased during operated lung collapse and hypoperfusion. At the same time (T1), other oxidative stress marker levels increased, including GSSG, oxidized glutathione/total glutathione ratio, GSSH–GSH couple redox potential, and MDA. This finding may be explained by the paradoxical cellular generation of ROS under conditions of ischaemia or lowered oxygen, using the considerable residual O₂ present during ischaemia. In such situations, the respiratory cytochromes become redox-reduced, enabling direct electron transfer to oxygen, producing large amounts of superoxide anions. Mitochondria-generated ROS apparently function as second messengers during hypoxia, contributing to signal transduction and leading to smooth muscle cell contraction in hypoxic pulmonary vasoconstriction. The concept of ischaemia-generated ROS is highly important, as it may contribute to oxidative cellular injury and play a signalling role during preconditioning.^{27282930–31}

All measured values showed larger increases immediately after resuming TLV by reperfusion phenomena at T2. Studies in other organs support the notion of ROS-generated reperfusion injury upon oxygen reintroduction to ischaemic tissue.^30,32,33 Our results provide the first evidence of significantly increased glutathione oxidation in this area, which correlates with an imbalance in the glutathione redox status (oxidized glutathione/total glutathione ratio). This suggests increased oxidative stress during and after OLV, probably due to multi-factorial causes, including lung collapse/hypoperfusion, tissue manipulation, and one-lung mechanical ventilation.

Our study showed that all oxidative injury marker blood levels were directly positively correlated with total lung collapse duration, and were always higher in T2 than T1. However, the decreased GSH levels were not correlated with OLV duration, possibly due to excessive GSH levels relative to the GSSG levels. These results are consistent with the inverse correlation between arterial blood PO₂/FiO₂ ratios and the levels of GSSG, oxidized glutathione/total glutathione ratio, and MDA before and after resuming TLV from OLV. Our findings suggest altered oxygen transport in the operated lung, which is related to diffuse alveolar damage induced at least partly by ROS and its contribution to gas exchange impairment after thoracic surgery. At T2, the whole blood GSSH–GSH couple redox potentials showed increased oxidation levels, in parallel with whole blood GSH and GSSG levels. Moreover, total lung collapse duration was inversely correlated with arterial blood PO₂/FiO₂, particularly at T1 (during operated lung collapse) and T4 (possibly due to impaired oxygen transport in the operated lung). Redox potential assessment is a promising tool for investigating both normal and pathological situations, including congenital mitochondrial diseases and in vitro studies of proliferation, differentiation, apoptosis, and cellular necrosis.^21,22,34,35

Importantly, our present results showed that total lung collapse duration during lobectomy was significantly correlated with glutathione oxidation and other oxidative damage marker levels. Our findings support the hypothesis of a direct relationship between OLV duration and the severity of oxidative damage during reperfusion in the expanded lung. Cancer patients undergoing lung lobectomy experience increased ROS formation, with increased glutathione oxidation and other oxidative stress marker levels in blood during OLV. These increases were larger after resuming TLV and were correlated with lung collapse and OLV duration. Furthermore, glutathione oxidation and lipid peroxidation were inversely correlated with arterial blood PO₂/FiO₂ values during and after OLV, possibly due to alveolar hypoxia and hypoxaemia before and after resuming TLV.

Disclaimer statements

Contributors J.G.D.L.A. takes responsibility for the integrity of the data. Study concept and design: J.G.D.L.A. Acquisition of data: E.G.D.O., G.G., F.M., R.B., J.P.G., A.D., C.D., J.C. Analysis and interpretation of data: J.G.D.L.A., E.G.D.O., F.M., J.P.G. Drafting of the manuscript: J.G.D.L.A., E.G.D.O. Critical revision of the manuscript: J.G.D.L.A., J.B., G.G., R.G., R.B. Statistical analysis: F.M. Administrative, technical, or material support: A.D., C.D., J.C. Study supervision: J.G.D.L.A.

Conflict of interest None declared.

Ethics approval The Ethic Committee at Hospital Clínico Universitario de Valencia (2006/184) approved the protocol and an informed consent was obtained from each patient.

Acknowledgments

The authors thank Prof Dr José Viña and Prof Dr Federico Pallardó (Department of Physiology, University of Valencia) for their generous help with GSH and GSSG determination.

Additional information

Funding

This work was supported in part by a grant (PI07/0836) from Instituto de Salud Carlos III to José García-de-la-Asunción.