|Year : 2019 | Volume
| Issue : 3 | Page : 583-588
Evaluation of arterial to end-tidal carbon dioxide pressure differences during laparoscopic renal surgery in the lateral decubitus position
Shalvi Mahajan1, Rajeev Chauhan2, Ankur Luthra2, Indu Bala2, Neerja Bharti2, Ashish Sharma3
1 Department of Anaesthesia and Intensive Care, Fortis Hospital Mohali, Chandigarh, India
2 Department of Anaesthesia and Intensive Care, PGIMER, Chandigarh, India
3 Department of Renal Transplant Surgery, PGIMER, Chandigarh, India
|Date of Web Publication||20-Sep-2019|
Department of Anaesthesia and Intensive Care, PGIMER, Chandigarh
Source of Support: None, Conflict of Interest: None
| Abstract|| |
Background: End-tidal carbon dioxide (PEtCO2) is a noninvasive reliable technique to measure arterial partial pressure of carbon dioxide (PaCO2) in the body under general anesthesia. However, gradient between PaCO2and PEtCO2(P[a-Et] CO2) is influenced by many factors. Aims: In the present study, we evaluated the changes in P (a-Et) CO2for laparoscopic donor nephrectomy in lateral decubitus position (LDP). Settings and Design: This was an observational, double-blinded, tertiary care center-based study. Methods: Thirty-one American Society of Anesthesiologists Class I and Class II patients of either sex undergoing laparoscopic donor nephrectomy in LDP under general anesthesia were included. An arterial cannula was inserted, PaCO2was measured at eight predesignated time intervals, and PEtCO2was also noted at the corresponding time period. Statistical Analysis: Data were analyzed using a two-way analysis of variance for repeated measurements using one dependent variable and one within-subject factor (time). Quantitative data were presented as mean ± standard deviation or median and interquartile range, as appropriate. Results: The mean P (a-Et) CO2gradient was 5.67 ± 1.36 mmHg 10 min after induction of anesthesia in the supine position (T1a). Ten minutes after LDP, P (a-Et) CO2gradient was 7.38 ± 1.45 mmHg (T1b) and was higher than T1a. The P (a-Et) CO2values 10 min after release of pneumoperitoneum and 10 min after making the patient supine were significantly higher than the T1a value. The highest value of P (a-Et) CO2gradient was at 30 min after creation of pneumoperitoneum (T30), i.e., 9.99 ± 1.70 mmHg. Pearson's correlation coefficient showed that the degree of correlation varied considerably during surgery due to interindividual variability (R2 T1a vs. T60 was 0.61 vs. 0.17). Conclusions: PEtCO2does not reliably predict PaCO2in healthy patients scheduled for laparoscopic renal surgery in LDP.
Keywords: Carbon dioxide, laparoscopic surgery, lateral decubitus position, partial pressure, pneumoperitoneum
|How to cite this article:|
Mahajan S, Chauhan R, Luthra A, Bala I, Bharti N, Sharma A. Evaluation of arterial to end-tidal carbon dioxide pressure differences during laparoscopic renal surgery in the lateral decubitus position. Anesth Essays Res 2019;13:583-8
|How to cite this URL:|
Mahajan S, Chauhan R, Luthra A, Bala I, Bharti N, Sharma A. Evaluation of arterial to end-tidal carbon dioxide pressure differences during laparoscopic renal surgery in the lateral decubitus position. Anesth Essays Res [serial online] 2019 [cited 2021 May 10];13:583-8. Available from: https://www.aeronline.org/text.asp?2019/13/3/583/264616
| Introduction|| |
End-tidal carbon dioxide (PEtCO2) monitoring is a well-established, simple, and noninvasive means of assessing alveolar ventilation. In contrary, arterial blood gas (ABG) measurement is regarded as the gold standard technique for assessment of partial arterial pressure of carbon dioxide (PaCO2). ABG measurement is an invasive, expensive technique and provides only intermittent values.
Laparoscopic renal surgery is performed with the patient lying in the lateral decubitus position (LDP). LDP is known to impair ventilation/perfusion (V/Q) relationship in anesthetized patients due to weight of mediastinum and disproportionate cephalad pressure of abdominal contents on dependent lung resulting in overventilation of nondependent lung. The pulmonary blood flow to dependent lung increases owing to effect of gravity leading to ventilation compromise. Further, creation of pneumoperitoneum during laparoscopy decreases functional residual capacity (FRC) and thoracic compliance and increases CO2 load in the body., Therefore, we hypothesized that patients undergoing laparoscopic renal surgeries in LDP may show significant variations inP (a-Et) CO2 in comparison to the LDP alone. However, no study has been carried out to determine these changes in the past. Hence, the present study was planned to evaluate the changes inP (a-Et) CO2 in patients scheduled for laparoscopic renal surgery in LDP under general anesthesia.
| Methods|| |
Thirty-one patients of the American Society of Anesthesiologists physical status (ASA-PS) Class I and Class II of either sex aged between 18 and 60 years undergoing elective laparoscopic renal surgery were included in this observational study.
The patients having pulmonary disease (chronic smoker), cardiovascular disease, intracranial pathology with raised intracranial pressure, obesity with body mass index >30 kg/m2, laparoscopy replaced by laparotomy and bilateral abnormal modified Allen's test (delayed hand flushing >10 s) were excluded from the study. Approval from the institute ethics committee and written informed consent were obtained from all the patients.
All patients underwent detailed preoperative assessment a day before surgery. Routine investigations such as hemoglobin, serum electrolytes, renal function tests, liver function tests, chest X-ray, electrocardiogram (ECG), and coagulogram were done in all patients. Modified Allen's test was done to assess the adequacy of collateral blood flow in both the hands. Patients were kept fasting for 8 h for solid food and 2 h for clear fluids and received premedication with tablet alprazolam 0.25 mg, the night prior to surgery and in the morning of surgery.
A standardized anesthesia protocol was followed. Standard monitoring such as 5-lead ECG, noninvasive blood pressure, and pulse oximeter (SpO2) was applied. Intravenous (i.v.) access was established, and infusion of normal saline at 8 mL/kg/h was started. General anesthesia was induced using i.v. thiopentone 4–5 mg.kg-1 and vecuronium 0.1 mg.kg-1 to facilitate tracheal intubation. Morphine 0.1 mg.kg-1 was administered for intraoperative analgesia. Postinduction arterial cannula was placed under all aseptic conditions in the dependent hand for continuous invasive blood pressure monitoring and blood sampling. If the modified Allen's test was abnormal in the dependent hand, arterial cannula was placed in the nondependent hand. Anesthesia was maintained using 60% nitrous oxide in oxygen (O2), isoflurane 1%–1.5% (minimum alveolar concentration: 1.2), and vecuronium given intermittently. All patients were mechanically ventilated initially with tidal volume 8–10 mg.kg-1 and inspiratory: expiratory time ratio 1:2 and respiratory rate (RR) of 12 breaths/min. During the surgery, RR was further adjusted to maintain PEtCO2 between 32 and 35 mmHg as measured by anesthesia gas analyzers. To facilitate the surgical procedure, patients were placed on nonoperative side in LDP. Pneumoperitoneum was established with CO2 insufflation at the rate of 2.5 L/min. Intra-abdominal pressure (IAP) was maintained at 12–14 mmHg throughout the surgical procedure. At the end of anesthesia, residual neuromuscular blockade was reversed by administration of glycopyrrolate 10 μg/kg and neostigmine 50 μg/kg. Trachea was extubated when a patient was awake and responded to verbal commands.
PaCO2 was measured using arterial blood (2 mL) which was immediately analyzed in the operation theater complex laboratory. Blood gas analysis was performed at 37°C by Siemens 855 (R-855) blood gas analyzer, and the results were automatically corrected to the patient's temperature.
PEtCO2 was measured by anesthesia gas analyzer of Datex Ohmeda S5 Avance workstation. Samples were taken from the manifold located between Y piece and elbow of the disposable anesthesia circuit. The mean PEtCO2 was determined by averaging the values obtained during the 15 s immediately before collecting arterial blood sample.
Intraoperative monitoring and data collection – systolic, diastolic, mean arterial pressure (SAP, DAP, and MAP), heart rate (HR), SpO2, PEtCO2, PaCO2, and temperature – were monitored and compared at the following stages: Stage I (before pneumoperitoneum) – T1a: ten minutes (min) after induction of anesthesia when hemodynamic status of a patient was stabilized and T1b: ten min after placing the patient in LDP and prior to CO2 insufflation; Stage II (during pneumoperitoneum): during the period of pneumoperitoneum at 30 min, 60 min, and 90 min and labeled as T30, T60, and T120; and Stage III (after release of pneumoperitoneum) – T2a: ten min after release of CO2 and T2b: ten min after a patient was made supine at the end of procedure.
Assuming a mean difference of 3.1 and standard deviation of 3.5 inP (a-Et) CO2 during general anesthesia in the supine position and in LDP, our sample size came out to be 28 and power of 90% and confidence interval of 95%. For the possible dropouts, it was decided to include 31 patients. Hemodynamic data and differences inP (a-Et) CO2 values were analyzed with a two-way analysis of variance for repeated measures. A Bland–Altman analysis was plotted for the correlation betweenP (a-Et) CO2 values, and a corresponding Spearman's rank correlation coefficient was computed for each of designated time points. P < 0.05 was considered clinically significant. All calculations were two-sided and performed using SPSS version 25 (Statistical Packages for the Social Sciences, Chicago, IL, USA).
| Results|| |
The present study was conducted in 31 patients who underwent laparoscopic donor nephrectomy in the LDP. There were 25 females and 6 males. Most of the patients belonged to ASA-PS Class I (28) and the remaining three belonged to ASA-PS Class II. Surgical procedure and demographic data are shown in [Table 1]. Eight concurrent sets of hemodynamic parameters (HR, SAP, DAP, and MAP); temperature; PEtCO2; and PaCO2 were recorded at predesignated time points [Table 2].
- PEtCO2: The mean PEtCO2 value recorded 10 min postinduction was 32.48 ± 2.12 mmHg (T1a), and a significant increase was observed in mean PEtCO2 at T1b, T30/T60/T120, T2a, and T2b. A significant rise in mean PEtCO2 values as compared with LDP (T1b) was observed at two points – T30 and T2b. The highest value of PEtCO2 was observed at 30 min after creation of pneumoperitoneum (T30). A rise in PEtCO2 value was observed after placing patient supine (T2b), and the values were higher than those observed after release of pneumoperitoneum (T2a) [Figure 1]a
- Arterial PaCO2: The mean PaCO2 was 38.16 ± 2.64 mmHg postinduction of anesthesia after stabilization of hemodynamics (T1a), and a significant increase in PaCO2 was observed at T1b, T30, T60, T120, and at T2a time points. Similarly, PaCO2 values were significantly higher at T30 to T120 time points and at T2a time point in comparison to the values observed after LDP (T1b). Thirty min after creation of pneumoperitoneum, the PaCO2 values was 44.20 ± 2.39 mmHg (range: 40.6–48.9 mmHg), and thereafter, a significant fall in values was observed 60–120 min. A further rise in PaCO2 value was observed 10 min after making supine position (T2b) as compared with the values observed after release of pneumoperitoneum (T2a) [Figure 1]b
- Arterial to end-tidal carbon dioxide gradient (P[a-Et] CO2): The meanP (a-Et) CO2 gradient was 5.7 ± 1.37 mmHg at T1a point (postinduction in supine) which showed a significant increase to 7.38 ± 1.45 mmHg 10 min after placing the patient in LDP (T1b). Thereafter, theP (a-Et) CO2 values remained significantly higher than the 10 min postinduction values throughout the study period (T1a). TheP (a-Et) CO2 values were significantly higher 30–120 min after creation of pneumoperitoneum in comparison to the values observed 10 min after LDP (T1b). The highestP (a-Et) CO2 value was recorded at 30 min after creation of pneumoperitoneum (9.99 ± 1.70 mmHg). Further decrease inP (a-Et) CO2 gradient was observed thereafter, and the values were significantly lower at 60 and 120 min and after the release of pneumoperitoneum [Figure 1]c.
|Figure 1: (a) End-tidal carbon dioxide at predesignated time intervals. (b) Partial arterial pressure of carbon dioxide at predesignated time intervals. (c) Arterial to end-tidal carbon dioxide gradient (P[aEt] CO2) at predesignated time intervals|
Click here to view
Pearson's correlation coefficient was calculated to assess the correlation between PEtCO2 and PaCO2 at different time intervals. The degree of correlation varied considerably before, during, and after pneumoperitoneum (R2 = 0.61 at 10 min postinduction, R2 = 0.17 at 60 min after pneumoperitoneum), as shown in [Table 3]. Throughout the procedure, the correlation between PEtCO2 and PaCO2 was positive, but due to interpatient variability, the correlation coefficient varied considerably. [Figure 2] shows Bland–Altman plots to demonstrate agreement between two methods of measuring CO2 partial pressure. The calculated mean difference and standard deviation and 95% limits of agreement are presented.
|Table 3: Correlation between end-tidal carbon dioxide and partial arterial pressure of carbon dioxide|
Click here to view
|Figure 2: Agreement between end-tidal carbon dioxide and partial arterial pressure of carbon dioxide at predesignated time intervals (Bland–Altman plot). Time interval: T1a-10 minutes post induction supine position. T1b-10 minutes after Lateral Decubitus Position. T30-30 minutes after creation of pneumoperitoneum. T60-60 minutes after creation of pneumoperitoneum. T120-120 minutes after creation of pneumoperitoneum. T2a-10 minutes after release of pneumoperitoneum. T2b-10 minutes after supine position|
Click here to view
| Discussion|| |
Monitoring of PEtCO2 is accepted as the standard of care to assess the adequacy of ventilation during the intraoperative period in patients undergoing general anesthesia and is considered to be clinically invaluable in providing immediate information of CO2 production, V/Q status, and elimination of CO2 from lungs. Although ABG measurement of PaCO2 is regarded as the gold standard technique, it provides only a snapshot of a continuously changing variable.
Although there is a good correlation between arterial and end-tidal CO2, several studies report significant variations within individual patients, especially in cardiac and aortic surgeries,, or laparoscopy. Continuous PEtCO2 monitoring is found as a reliable indicator of the trend in arterial CO2 fluctuations in the ASA-PS Class I and Class II patients undergoing laparoscopic nephrectomy under general anesthesia in the supine position.
In the present study, meanP (a-Et) CO2 gradient 10 min postinduction was 5.67 ± 1.36 mmHg, with a range of 4–11.5 mmHg. This correlates reasonably well with the values reported earlier. In previous studies, meanP (a-Et) CO2 gradient of 4–6 mmHg has been noted in anesthetized, normothermic patients who are lying in the supine position.
TheP (a-Et) CO2 gradient recorded 10 min after LDP was 7.38 ± 1.45 mmHg and was significantly higher than the value recorded in the supine position (5.67 ± 1.36 mmHg). Both PEtCO2 and PaCO2 showed a significant rise from the supine position. Grenier et al. recorded a significantly higherP (a-Et) CO2 gradient in neurosurgical patients operated in LDP in compared to those who were operated in the supine position (7 ± 3 mmHg vs. 6 ± 3 mmHg in lateral vs. supine position, P < 0.05). Similarly, Pansard et al. found a significant increase inP (a-Et) CO2 gradient 5 min after LDP in comparison to the supine position (7.9 ± 3.5 mmHg vs. 4.8 ± 3.9 mmHg in lateral vs. supine position P < 0.05) in patients undergoing renal surgery in the LDP. However, contrary to our findings, no change in PaCO2 was observed, whereas PEtCO2 showed a significant decrease in LDP in comparison to supine position. This difference from our results could be because of raised kidney bridge in the study of Pansard et al. which could have caused more hemodynamic alterations with a decrease in right-sided filling pressure or cardiac output or both even though MAP did not show significant change. The enlargement ofP (a-Et) CO2 gradient observed after positioning in LDP could be related to alterations in V/Q relationship that are known to occur in LDP. The anesthetized patient in LDP has a nondependent lung that is well ventilated but poorly perfused, whereas the dependent lung is well perfused but poorly ventilated leading to an increased degree of V/Q mismatch. Thus, physiological dead space increases leading to an enlargedP (a-Et) CO2 gradient in LDP.
It was observed that an increase in P (a-Et) CO2 gradient from 30 min to 120 min after creation of pneumoperitoneum and this increase was significant when compared to the values after induction and lateral decubitus position. Parikh et al. conducted a study in patients undergoing laparoscopic kidney transplant surgery in steep Trendelenburg position and observed a mild rise inP (a-Et) CO2 gradient after creation of pneumoperitoneum which persisted throughout the procedure and even after extubation. However, the rise in gradient did not reach statistical significance. Laparoscopic procedures performed with CO2 pneumoperitoneum increase CO2 load by absorption of CO2 via peritoneal surface. There may also be a decrease in both thoracic compliance, and FRC., In majority of the studies, PaCO2, PEtCO2, andP (a-Et) CO2 have been found to increase and affected by duration of pneumoperitoneum and body position. In some reports, P (a-Et) CO2 gradient either did not change, or even a negative gradient has been reported in patients with impaired pulmonary function.
The highest value ofP (a-Et) CO2 gradient was observed 30 min after pneumoperitoneum (9.99 ± 1.70 mmHg) during the study. There are no previous reports of laparoscopic renal surgery in the LDP. During laparoscopic colorectal surgery in Trendelenburg and reverse Trendelenburg position, Tanaka et al. observed the highest value ofP (a-Et) CO2 gradient at 60 min, whereas Kim recorded at 120 min after pneumoperitoneum. The difference was attributed due to lower IAP (7–10 mmHg) as compared to higher IAP (12 mmHg) used in respective studies. In our study, the time to reach the highest value ofP (a-Et) CO2 was 30 min after pneumoperitoneum which is shorter than the time reported in laparoscopic colorectal surgeries. The reason for the shorter time to reach the highest value could be attributed to the combined effects of CO2 pneumoperitoneum and the increased degree of mismatching of V/Q and increased physiological dead space to the extent of 13% in the LDP.
Further, in our study, a significant decrease in PaCO2, PEtCO2, andP (a-Et) CO2 was seen at 60 and 120 min after pneumoperitoneum as compared to the values observed at 30 min though it was still significantly higher than the supine and LDP. This fall in PEtCO2 could be because of the repeated adjustment in minute ventilation to maintain PEtCO2 between a narrow range of 32–35 mmHg. The fall in PaCO2 values after 30 min of pneumoperitoneum may be due to limited capacity of the body to store CO2 and impaired local perfusion due to increased IAP which prevents further absorption of CO2.
We observed an increase in PEtCO2, PaCO2, andP (a-Et) CO2 gradient after making supine position postsurgery as compared to the values after of release of pneumoperitoneum and postinduction values. Hirvonen et al. also observed an increase in PaCO2 levels a few minutes after deflation of pneumoperitoneum. Following desufflation, local perfusion improves and absorption of CO2 across the resection surfaces increases. Carbon-di oxide gas, which has been accumulated in the collapsed peritoneal capillary vessels, reaches the systemic circulation leading to an increase in PaCO2. Also, increased dead space ventilation and increase in cardiac output following decrease in intrathoracic pressure increase PEtCO2 and PaCO2. In general, PaCO2 returns to normal range within 1 h of desufflation, but after prolonged laparoscopic surgery, it may take several hours to achieve a steady state of CO2 as a considerable amount of CO2 is stored in the extravascular compartments of the body gets slowly redistributed, metabolized, or exhaled.
Although the correlation between PEtCO2 and PaCO2 was positive throughout the study period, the degree of correlation at different time points varied considerably due to interpatient variability (R2 = 0.61 at 10 min postinduction and R2 = 0.17 at 60 min after PP).
Hemodynamic parameters remained stable throughout the study period. None of the patients had significant bradycardia or any vasopressor requirement so as to treat hypotension.
In the present study, we included only healthy patients without cardiopulmonary disease. Majority of the patients were female and belonged to the younger age group, with a mean age of 40.94 ± 9.94 years. Therefore, the gender, age range of patients, and selection of ASA-PS Class I and Class II patients may limit the generalization of our data to the population in the older age group and those with cardiopulmonary disease or abnormal pulmonary function.
| Conclusions|| |
PEtCO2 does not reliably predict PaCO2 in healthy patients scheduled for laparoscopic renal surgery performed in the LDP. We recommend that arterial sampling for measuring PaCO2 if available should be used during laparoscopic renal surgery to confirm the adequacy of ventilation.
Financial support and sponsorship
Conflicts of interest
There are no conflicts of interest.
| References|| |
Benallal H, Busso T. Analysis of end-tidal and arterial PCO2 gradients using a breathing model. Eur J Appl Physiol 2000;83:402-8.
Cohen E. Physiology of the lateral position and one-lung ventilation. Chest Surg Clin N
Bardoczky GI, Engelman E, Levarlet M, Simon P. Ventilatory effects of pneumoperitoneum monitored with continuous spirometry. Anaesthesia 1993;48:309-11.
Fahy BG, Barnas GM, Nagle SE, Flowers JL, Njoku MJ, Agarwal M. Changes in lung and chest wall properties with abdominal insufflation of carbon dioxide are immediately reversible. Anesth Analg 1996;82:501-5.
Swaim JA. Hypothermia and blood pH. Arch Intern Med 1988;148:1643-6.
Raemer DB, Francis D, Philip JH, Gabel RA. Variation in PCO2 between arterial blood and peak expired gas during anesthesia. Anesth Analg 1983;62:1065-9.
Bermudez J, Lichtiger M. Increases in arterial to end-tidal CO2 tension differences after cardiopulmonary bypass. Anesth Analg 1987;66:690-2.
Brampton WJ, Watson RJ. Arterial to end-tidal carbon dioxide tension difference during laparoscopy. Magnitude and effect of anaesthetic technique. Anaesthesia 1990;45:210-4.
Jayan N, Jacob JS, Mathew M. Anaesthesia for laparoscopic nephrectomy: Does end-tidal carbon dioxide measurement correlate with arterial carbon dioxide measurement? Indian J Anaesth 2018;62:298-302.
] [Full text]
Nunn JF, Hill DW. Respiratory dead space and arterial to end-tidal CO2
tension difference in anesthetized man. Appl J Physiol 1960;15:383-9.
Grenier B, Verchère E, Mesli A, Dubreuil M, Siao D, Vandendriessche M. Capnography monitoring during neurosurgery: Reliability in relation to various intraoperative positions. Anesth Analg 1999;88:43-8.
Pansard JL, Cholley B, Devilliers C, Clergue F, Viars P. Variation in arterial to end-tidal CO2 tension differences during anesthesia in the “kidney rest” lateral decubitus position. Anesth Analg 1992;75:506-10.
Parikh BK, Shah VR, Modi PR, Butala BP, Parikh GP. Anaesthesia for laparoscopic kidney transplantation: Influence of trendelenburg position and CO2 pneumoperitoneum on cardiovascular, respiratory and renal function. Indian J Anaesth 2013;57:253-8.
] [Full text]
McMahon AJ, Fischbacher CM, Frame SH, MacLeod MC. Impact of laparoscopic cholecystectomy: A population-based study. Lancet 2000;356:1632-7.
Puri GD, Singh H. Ventilatory effects of laparoscopy under general anaesthesia. Br J Anaesth 1992;68:211-3.
Ciofolo MJ, Clergue F, Seebacher J, Lefebvre G, Viars P. Ventilatory effects of laparoscopy under epidural anesthesia. Anesth Analg 1990;70:357-61.
Joris J, Ledoux D, Honore P, Lamy M. Ventilatory effects of CO2 insufflation during laparoscopic cholecystectomy. Anesthesiology 1991;75:A121.
Wahba RW, Mamazza J. Ventilatory requirements during laparoscopic cholecystectomy. Can J Anaesth 1993;40:206-10.
Tanaka T, Satoh K, Torii Y, Suzuki M, Furutani H, Harioka T, et al.
Arterial to end-tidal carbon dioxide tension difference during laparoscopic colorectal surgery. Masui 2006;55:988-91.
Kim YS. Arterial and end-tidal carbon dioxide pressure differences during laparoscopic colorectal surgery. Eur J Anaesthesiol 2008;25:74-5.
Werner O, Malmkvist G, Beckman A, Stahle S, Nordström L. Carbon dioxide elimination from each lung during endobronchial anaesthesia. Effects of posture and pulmonary arterial pressure. Br J Anaesth 1984;56:995-1001.
Joris JL. Anesthesia for laparoscopic surgery. In: Miller RD, Eriksson LI, Fleisher LA, Wiener-Kronish JP, Young WL, editors. Miller's Anesthesia. Philadelphia: Churchill Livingstone; 2010. p. 2185-202.
Hirvonen EA, Nuutinen LS, Kauko M. Ventilatory effects, blood gas changes, and oxygen consumption during laparoscopic hysterectomy. Anesth Analg 1995;80:961-6.
[Figure 1], [Figure 2]
[Table 1], [Table 2], [Table 3]