Published ahead of print on May 1, 2008, doi:10.1164/rccm.200801-113OC
© 2008 American Thoracic Society doi: 10.1164/rccm.200801-113OC
Hypoxemia in Inferior Pulmonary Veins in Supine Position Is Dependent on Obesity1 Division of Cardiology, Department of Internal Medicine, Jikei University School of Medicine, Tokyo, Japan Correspondence and requests for reprints should be addressed to Teiichi Yamane, M.D., Division of Cardiology, Department of Internal Medicine, Jikei University School of Medicine, 3-25-8, Nishi-shinbashi, Minato-ku, Tokyo 105-8461, Japan. E-mail: yamanet1{at}aol.com
Rationale: Although arterial oxygen concentration decreases in obese subjects, the mechanism for this remains to be determined. Objectives: The blood gas level in each pulmonary vein (PV) was measured in supine subjects with diverse body mass index (BMI) values, to determine whether there was a regional insufficiency in gas exchange depending on the subject's BMI. Methods: Forty subjects with normal cardiopulmonary function who underwent a catheter ablation for atrial fibrillation were included. Before delivering any radiofrequency energy application, blood samples were obtained from each of the four PVs during physiologic breathing in a supine position to measure the PO2 and PCO2 values. Spirometry and ventilation/perfusion lung scintigraphy were also performed. Measurements and Main Results: The PO2 value was significantly higher in superior veins than inferior veins (91.8 ± 13.5, 70.8 ± 16.3, 92.2 ± 11.1, and 73.6 ± 13.7 mm Hg, in the left superior, left inferior, right superior, and right inferior PV, respectively). There was a significant inverse relationship between the PO2 and PCO2 values. Neither the spirometry nor lung scintigraphy could detect any remarkable findings corresponding to the low PO2 values. Among the various clinical characteristics, only the BMI was significantly associated with the decreased PO2 value in the inferior veins. Conclusions: Hypoxia in obese subjects in a supine position is thus considered to be primarily caused by insufficient gas exchange in the regions of lung linked to the inferior PVs. The inverse relationship between the BMI and PO2 value in the inferior PVs suggests a possible subclinical manifestation of obesity-related respiratory insufficiency.
Key Words: pulmonary vein • hypoxia • obesity • supine position
Respiratory complications associated with obesity, such as sleep apnea syndrome or obesity–hypoventilation syndrome, have been previously addressed in various studies (1–3). In normal subjects, there have been reports describing the association between an increased body mass index (BMI) and decreased arterial oxygen concentration (PaO2) (4, 5). Regardless of these studies, the true mechanism of hypoxia in obese subjects remains to be determined. This study analyzed the features of obesity-related respiratory insufficiency by evaluating the O2 and CO2 concentrations in each pulmonary vein (PV).
This study included 40 subjects with normal cardiac function (left ventricular ejection fraction [LVEF] > 50%) who underwent catheter ablation (PV isolation) for atrial fibrillation (AF) (paroxysmal, 33 subjects; persistent or chronic, 7 subjects). There were 35 men and 5 women with a mean age of 54 ± 10 years. The clinical characteristics of the patients are shown in Table 1. The chest radiography (X-P) was unremarkable in all cases with a mean cardiothoracic ratio (CTR) of 48.0 ± 5.9%. The BMI value ranged from 17.5 to 30.0 (mean, 23.2 ± 2.7). In all cases, the absence of any anomalous pulmonary venous return was confirmed by a preoperative multidetector computed tomography scan.
Written, informed consent was obtained from all patients before the procedure according to the protocol approved by Jikei University Hospital's human research committee. All catheters were inserted through the femoral approach except for one catheter inserted into the coronary sinus through the subclavian approach. The left atrium (LA) and PVs were explored through either a patent foramen ovale (5 patients) or transseptal puncture (Brockenbrough method). The direct visualization of all four PVs was first performed using selective venography through a long sheath (8 F, SR0; St. Jude Medical, Minnetonka, MN) for the superior PVs or using a 6-F NIH catheter (Cathex; Cathex, Kanagawa, Japan) for the inferior PVs, with a hand injection of contrast medium (Figure 1).
Before delivering any radiofrequency application, blood samples were obtained from each of the four PVs and femoral artery. The blood samples (  2 ml) were drawn either through a 6-F NIH catheter in the LA or PVs or through a 4-F guiding sheath placed in the femoral artery. Special care was taken to discard blood that had been lying in the catheter; and blood samples were then slowly obtained from deep inside each PV to avoid contamination with the atrial blood (Figure 1). PO2 and PCO2 values were measured within 10 minutes of their sampling by a blood gas analyzer (ABL715; Radiometer, Copenhagen, Denmark). Two-point and one-point calibrations were set up every 8 and 4 hours, respectively. Neither sedative agents nor oxygen was given to the subjects at any time before the end of taking all the blood samples.
Ventilation function tests (spirometry: CHESTAC-9800; Chest Co., Tokyo, Japan) and pulmonary ventilation/perfusion ( The values are expressed as the mean ± SD. Student t test (paired) was used to compare the paired PO2 or PCO2 values between the ipsilateral superior and inferior PVs. A Pearson's correlation coefficient test was used to evaluate the correlation between two variables (univariate regression analysis). A multiple linear regression analysis was used to determine factors independently associated with dependent variables (mean PO2 value of inferior PVs). Covariates in this model were BMI, age, sex, smoking, AF type, CTR, LA dimension, and LVEF. No statistically significant interactions were observed among the predictor variables in this model. The absence of severe multicollinearity among variables was also confirmed by the following: (1) there were no pairs of predictor variables with a strong correlation (the highest correlation observed was 0.47 between the BMI and CTR) and (2) there were no predictor variables with a high value of variance inflation factor (range from 1.18 to 1.98). Statistical analyses were performed with the use of a statistical software program (SPSS for Windows, version 11.5; SPSS, Inc., Chicago, IL). Differences with a P < 0.05 were considered to be statistically significant.
The partial pressures of oxygen in each of the four PVs and in the femoral artery under spontaneous breathing of room air in all cases studied are shown in Figure 2A and Table 1. Twenty-seven cases were in sinus rhythm during the blood sampling, whereas the remaining 13 were in AF. Although all cases had normal cardiopulmonary function, the PO2 value in the PVs varied markedly, from 43.2 to 126.0 mm Hg. The PO2 was significantly higher in the superior PVs than in the inferior PVs (mean value: 91.8 ± 13.5, 70.8 ± 16.3, 92.2 ± 11.1, and 73.6 ± 13.7 mm Hg, in the left superior, left inferior, right superior, and right inferior PVs, respectively; P < 0.01). PO2 values of less than 60 mm Hg were observed in 21 PVs in 15 cases, 95% (20/21 PVs) of which were identified in the inferior PVs. The mean difference between the highest and the lowest PO2 value in each case was 30.6 ± 11.3 mm Hg (8.2 to 57.1 mm Hg).
In contrast to the PO2 values, the PCO2 values in the same blood samples exhibited the opposite results, that is the PCO2 values in the inferior PVs were significantly higher than those of the ipsilateral superior PVs (mean value: 42.2 ± 4.5, 44.4 ± 4.2, 43.0 ± 3.6, and 44.4 ± 3.3 mm Hg, in the left superior, left inferior, right superior, and right inferior PVs, respectively; P < 0.05). Accordingly, a significant inverse relationship was observed between the PO2 and PCO2 values in each PV (Figure 3; n = 160, r = –0.49, P < 0.001).
The results of spirometry testing were unremarkable in all 22 examined subjects, with a mean FEV1.0% value of 80.1 ± 4.8 (72.3 to 90.0) and a mean %VC of 107.5 ± 13.0 (84.8 to 130.2). The results of the DLCO values were higher than 70% of the predicted value in all but one subject (case number 11, 67.9%). The calculated mean DLCO value was 91.7 ± 16.9% (67.9 to 147.8%). No significant  / mismatch could be detected by scintigraphy, except for a tiny perfusion defect in the upper lobe of the right lung, detected in two patients (case numbers 33 and 39). Figure 4 demonstrates the results of a univariate regression analysis undertaken between the BMI and the PO2 or PCO2 values from the femoral artery (Figures 4A and 4B), superior PVs (Figures 4C and 4D), and inferior PVs (Figures 4E and 4F). As shown in Figures 4A and 4B, the arterial PaO2 decreased proportionally as the BMI increased (R = –0.41, P = 0.0095), whereas PaCO2 values increased in line with the increased value of BMI (R = 0.31, P = 0.048). Figures 4C and 4D demonstrate that the mean PO2 and PCO2 values of both superior PVs had no significant correlation to the BMI value. In contrast, the mean PO2 value of both inferior PVs exhibited a significant inverse correlation to the BMI value (R = –0.57, P = 0.00011), and the mean PCO2 value of both inferior PVs increased according to the BMI value (R = 0.41, P = 0.0083; Figures 4E and 4F). It is notable that the above relationship between the blood gas values and BMI was stronger in the inferior PVs (Figures 4E and 4F) in comparison to that observed in the arterial blood (Figures 4A and 4B).
To identify the clinical factors related to the decreased oxygen tension in the inferior PVs other than the BMI, a multiple linear regression analysis was performed (Table 2). Among the various clinical factors (age, sex, smoking habit, type of AF, BMI, CTR on chest radiography, LA dimension, and LVEF), there were no clinical factors observed to be correlated to the decreased PO2 value in the inferior PVs except for the BMI (coefficient: –0.55, P = 0.005; 95% confidence interval, –4.40 to –0.88).
This study revealed several new findings. First, the oxygen tension varied greatly between the superior and inferior PVs during physiologic breathing in supine subjects without any serious cardiopulmonary disease. The PO2 value was inversely correlated to the PCO2 value. Second, both the arterial PaO2 and PaCO2 significantly correlated with the BMI value. With more regional analysis, the PO2 and PCO2 in the inferior PVs had a stronger correlation to the BMI value, but no significant correlation was observed in the superior PVs. This is the first systematic study to demonstrate the relationship between the BMI and regional hypoxemia in the lung. Although the efficiency of blood oxygenation differs depending on the region of the lungs, especially between the upper and lower portions of the lungs due to the effect of gravity (6), there have been few studies that measured the actual value of the oxygen concentration in each PV under physiologic conditions. The current data are unique in that they were obtained during physiologic breathing in supine subjects with normal cardiopulmonary function.
There have been a number of articles describing the influence of obesity and body position on pulmonary function (7–10). The most common pulmonary function abnormality in obese subjects is known to be a reduction in the expiratory reserve volume (ERV) and FRC (1, 7, 9). The ERV reduction is shown to be greatest in the supine position when the diaphragm ascends in the chest and the weight of the lower thorax and the abdomen is applied to the lungs. In an early report by Craig and colleagues, it was shown that the relationship between the FRC and closing volume (CV) differs according to body position (7). Because FRC fell in the supine position, whereas CV was unchanged, CV exceeded the FRC in some cases with the supine position, which resulted in airway closures and gas-trapping in the chest. Furthermore, Farebrother and coworkers (9) demonstrated that the arterial oxygen pressure was significantly less in the supine than in the sitting posture and improved after weight loss in obese subjects. The reduction in PaO2 was correlated with the extent to which dependent airways were closed within the range of tidal breathing (shown by increasing negativity of ERV-CV as a percentage of VC). Putting these together, the hypoxemia in the inferior PVs in our supine subjects can be attributed in part to the regional alveolar Obesity has been demonstrated to have many detrimental effects on the respiratory function, which may lead to chronic hypoventilation in subjects with profound obesity (1), known as obesity–hypoventilation syndrome (2, 3). In normal subjects, there have also been reports describing an association between the increased BMI and decreased PaO2 value (4, 5). The present study identified decreased PO2 and increased PCO2 values in the inferior PVs correlated with increased BMI value, as the underlying mechanism of the decreased arterial oxygen concentration in supine subjects. It is of special interest that these correlations between the PO2 (PCO2) and BMI were observed in subjects with a wide variety of BMI values, and not only in subjects with profound obesity, thus suggesting that the insufficient gas exchange in the inferior lung field could start at an early phase of obesity.
Study Limitations
Conclusions
The authors thank Dr. Futoshi Kotajima (Department of General Medicine, Jikei University Hospital) for his advice regarding respiratory physiology, Dr. Masato Matsushima (Department of General Medicine, Jikei University Hospital) for his advice regarding statistical analysis, and Dr. Brian Quinn (Japan Medical Communication) and Mr. John Martin (St. Jude Medical) for their kind advice on English-language use.
Originally Published in Press as DOI: 10.1164/rccm.200801-113OC on May 1, 2008 Conflict of Interest Statement: None of the authors has a financial relationship with a commercial entity that has an interest in the subject of this manuscript. Received in original form January 17, 2008; accepted in final form April 29, 2008
This article has been cited by other articles:
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
TOP
ABSTRACT
AT A GLANCE COMMENTARY
