Effect of lung volume reduction surgery on pulmonary hemodynamics in severe pulmonary emphysema 1

Objective: The presence of pulmonary hypertension in severe pulmonary emphysema has been considered a relative contraindi-cation to lung volume reduction surgery (LVRS). There was concern that resection of lung tissue might further increase pulmonary artery pressure. To address this point, the prevalence of pulmonary hypertension in candidates for LVRS was investigated. The changes in pulmonary artery pressures after bilateral videoassisted thoracoscopic resection was studied in patients with homo- and heterogeneously destroyed emphysematous lungs. Design: The pulmonary arterial pressures by right heart catheterization were prospectively assessed, before and 6 months after LVRS in 21 consecutive patients (15 males, six females, mean ( 9 S.E.) age: 62 9 1.9, range 42–74 years). All were former smokers and three had ZZ-AT1 deﬁciency. The inclusion criteria were: (a) severe bronchial obstruction (FEV 1 B 35% predicted); (b) pulmonary hyperinﬂation (RV : TLC \ 0.60); and (c) absence of hypercapnia (PaCO 2 B 50 mmHg). Results: The FEV 1 had increased from 28 9 2% to 35 9 3% of the predicted value ( P B 0.05) 6 months after surgery. The RV : TLC had declined from 0.65 9 0.02 to 0.55 9 0.02; PaO 2 increased (66 9 1 versus 71 9 2 mmHg, P (cid:30) 0.04), PaCO 2 (38 9 2 versus 36 9 1 mmHg, P (cid:30) 0.26) did not change. The pulmonary artery mean pressure (PAP mean ) remained unchanged (18 9 1 versus 19 9 1 mmHg, P (cid:30) 0.26). In six patients PAP mean was ] 20 mmHg (up to 24 mmHg) preoperatively. After 6 months, six patients had a PAP mean ] 20 mmHg (up to 31 mmHg). Conclusions: In patients with severe emphysema who are candidates for LVRS (but have only mild to moderate hypoxemia and a PaCO 2 B 50 mmHg) we found no relevant pulmonary hypertension and pulmonary artery pressure did not change signiﬁcantly after surgery. Therefore, routine right heart catheterization is not mandatory for preoperative evaluation. © 1998 Elsevier Science B.V. All rights reserved.


Introduction
Lung volume reduction surgery (LVRS) is a novel therapeutic concept which has been successfully applied in patients with airflow obstruction and hyperinflation due to severe pulmonary emphysema-who are severely limited in their daily activity [1][2][3][4]. LVRS is performed by median sternotomy [1,2] or by video-assisted thoracoscopy (VAT) [3][4][5]. For both types of procedures, the patient has to undergo one-lung ventilation. The intra-and postoperative management is a challenge in these patients, who are severely impaired in their pulmonary function [6]. Pulmonary arterial hypertension may be a risk for pre-and postoperative complications. Furthermore, there has been concern, that the resection of pulmonary vascular bed during LVRS might compromise pulmonary hemodynamics and hence functional improvement. Information on changes in pulmonary hemodynamics and right heart function after LVRS is scarce. Sciurba et al. [7] found an increase of the fractional change in right ventricular area after LVRS by echocardiography. In a preliminary study Kessler et al. [8] measured normal PAP mean at rest in four emphysema patients before surgery. It was found that hemodynamics at rest remained unchanged after LVRS and that the exercise induced rise in PAP mean was even less after surgery.
The goal of this prospective study was to assess the prevalence of pulmonary arterial hypertension in patients with severe pulmonary emphysema -who were selected for LVRS according to published criteria [4]and to study the effects of LVRS on pulmonary artery pressures 6 months after the procedure.

Inclusion criteria
According to criteria published previously [1,4] potential candidates for LVRS had the following profile: the patient has severe COPD with an FEV 1 of less than 35% predicted and is considerably hyperinflated (residual lung volume\ 200%, total lung capacity\ 130% predicted). Radiological signs of pulmonary emphysema with flat diaphragms are present on conventional chest X-ray. The emphysema is confirmed on a high resolution CT scan. The patient is highly motivated and has stopped smoking for more than 6 months. No further improvements could be achieved by antiobstructive pharmacotherapy including systemic corticosteroids.

Exclusion criteria
Age \ 75 years, PaCO 2 \55 mmHg, diffusing capacity for carbon monoxide (DL CO single breath)B 20% predicted, bronchiectases, acute broncho -pulmonary infection, neoplastic disease with a life expectancy of less than 2 years and psychiatric disturbance. Patients with a previous Q-wave myocardial infarction and/or congestive heart failure were also not considered to be candidates for LVRS.

Cardiac e6aluation
A careful history seeking to identify symptoms of prior right heart failure was obtained. Patients were clinically examined for: (a) elevated jugular vein pressure; (b) ankle edema; and (c) roentgenographic signs of pulmonary hypertension, i.e. a diameter of the right descending pulmonary artery diameter greater than 1.6 cm on conventional p.a. chest X-ray [9] and electrocardiographic features of right ventricular hypertrophy, i.e. right axis deviation greater than +100°and R/S ratio in V 1 \ 1 [10].

Dyspnea score
Dyspnea was assessed by the Medical Research Council (MRC) dyspnea score [11], by which shortness of breath is rated from 0 to 4 according to an increase in symptom severity.

Functional assessments
Lung volumes were measured according to standard criteria [12,13] using the Sensor Medics 6200 Autobox ® (Yorba Linda, Ca USA). Results were expressed as the best values after inhalation of two puffs of salbutamol. Diffusing capacity for carbon monoxide was measured using the 6200 ® /Sensor Medics equipment which uses methane as inert tracer gas. The reference values were according to the ECCS [14]. The arterial blood gases were analyzed using an AVL 995-S (AVL ® Medical Instruments, Schaffhausen, Switzerland) and oxygen saturation (HbO 2 ) and COHb with a CO-oximeter (IL 482 ® , Radiometer, Copenhagen). Blood was drawn from a radial artery while the patient was sitting at rest and breathing room air.

Right heart catheterization
By the femoral approach, a Swan-Ganz catheter 7 F (Criti Cath ® , Ohmeda, Singapore) was introduced into the right pulmonary artery. The pressures were recorded in the: (a) right atrium; (b) right ventricle; (c) pulmonary artery; and (d) in pulmonary capillary wedge position. The pressures were measured at FRC and averaged over nine cycles. The zero reference level for vascular pressures was taken as 5 cm below the sternal angle using fluoroscopy while the patient lay supine. Right heart catheterization was performed during the preoperative evaluation period for baseline and 6 months later. The patients were not receiving supplemental oxygen and were awake and not sedated.

Surgical technique
LVRS was performed in all patients bilaterally by video-assisted thoracoscopy as described previously [4]. Resection was aimed at the most destroyed tissues previously identified by CT scans and perfusion scintigraphy. The excised pieces of lung had an estimated cumulative volume of approximately 20 -30% of the lung volume on each side. The staplers were not buttressed with xenopericard. Extubation was performed in the theater immediately at the end of operation.

Statistics
Pre-and postoperative values are expressed as means 9 S.E. and compared by paired t-test. A probability of less than 0.05 was considered significant. Analysis of variance followed by the Newmann -Keuls multiple comparisons procedure, where appropriate, was performed to detect significant differences in preoperative versus postoperative values within the same group and in values between groups. Table 1, the patients had the typical functional features of severe pulmonary emphysema. These consisted of severe obstruction to expiratory airflow and considerable hyperinflation. Their MRC dyspnea score was 3.590.1. Patients had only mild hypoxemia with a mean PaO 2 of 66 mmHg at rest (Table 1). In only three subjects, PaO 2 was lower than 60 mmHg, i.e. 57, 54 and 52 mmHg. None of the patients were considerably hypercapnic (PaCO 2 \ 50 mmHg) since this was an exclusion criterion for surgery.

As shown in
No patient had a history of right heart decompensation or clinical signs of right heart failure -such as ankle edema or an elevated jugular venous pressure. In the preoperative chest x-ray, the mean diameter of the right descending pulmonary artery was 1.290.1 cm and was smaller than 1.6 cm in all LVRS candidates. In the preoperative ECG, none of the patients had signs of right ventricular overload. The mean R-axis was 519 49°, the P-axis was 14 9 13°and the R/S ratio in V 1 was 0.69 0.3. A right bundle branch block was not documented in any of the patients.
The pulmonary artery pressures are shown in Table 2 and Fig. 1. There were six patients with a PAP mean of 20 mmHg or greater preoperatively but the PAP mean wasn't above 25 mmHg in any of the patients (Fig. 1). The six patients with a PAP mean above 25 mmHg did not differ from those with normal PAP mean in respect to: (a) age; (b) gender; (c) preoperative FEV 1 % predicted; (d) RV/TLC; (e) arterial PaO 2 ; or oxygen saturation; and (f) PaCO 2 .  All patients had an uncomplicated postoperative course. Clinically relevant improvements in pulmonary function tests in respect to airways obstruction and hyperinflation were found, 6 months after surgery ( Table 1).The patients were considerably less short of breath: their MRC dyspnea score declined from 3.59 0.1 to 1.99 0.2 (P B0.0001).
A mild improvement in the PaO 2 ,was noticed but there were no significant changes in PaCO 2 and in the alveolar-arterial oxygen partial pressure gradient ( Table 1). The lung volume reduction surgery did not change the pulmonary diffusing capacity for carbon monoxide ( Table 1). The mean pulmonary artery pressures remained unchanged (Table 2). In three patients with a preoperative PAP mean ]20 mmHg, PAP mean was lower than 20 mmHg after LVRS, whereas in three patients with a normal preoperative PAP mean higher pressures were found postoperatively. We were unable to explore the effect of LVRS on the postoperative diameter of the right descending pulmonary artery, since p.a. chest-x rays were not performed routinely after surgery. There is have no data on postoperative ECG alterations.

Discussion
It has been shown that bilateral lung volume reduction surgery improves lung function in selected patients with severe emphysema. However, the preoperative predictors of operative morbidity and mortality are not yet well defined in this population. Since pulmonary hypertension might increase the preoperative risk and compromise the postoperative functional improvement, the prevalence of pulmonary hypertension was assessed.
Also studied were the changes in pulmonary artery pressures and gas exchange after videoassisted bilateral thoracoscopic LVRS in patients selected for LVRS according to published criteria [4].
The degree of changes in dyspnea and lung function that were observed 6 months after LVRS were comparable to the results of other groups, who perform this type of surgery bilaterally [1,4,15].
The PAP mean at rest was not elevated in the patient population and only six patients had mild pulmonary hypertension (PAP mean ] 20B 25 mmHg). These results are consistent with the findings of other investigators. They have previously shown, that pulmonary artery pressures are normal or only mildly elevated in patients with COPD and emphysema as long as significant hypoxemia and hypercapnia are not present [16,17]. In a population of COPD patients with a broad range of blood gas abnormalities, a negative correlation between resting pulmonary artery pressures and arterial oxygen saturation [18] as well as a positive correlation between the arterial PaCO 2 and pulmonary artery pressures are found [18,19]. The observation, that we were unable to find such a correlation in our patients is most likely due to the fact, that according to the selection criteria for LVRS the range of blood gases was small and only three of the patients showed mild hypoxemia. Based on these findings it is concluded, that in candidates for LVRS, who have only mild to moderate hypoxemia and a PaCO 2 B 50 mmHg, no relevant pulmonary hypertension is found. The routine right heart catheterization is therefore not mandatory for preoperative evaluation.
The mean pulmonary artery pressures remained unchanged after LVRS ( Fig. 1 and Table 1). In a few patients the PAP mean even declined after LVRS-in some the pulmonary artery pressures were somewhat higher after surgery. However, in none of the patients the postoperative PAP mean was higher than 35 mmHg, which is considered as relevant degree of secondary pulmonary hypertension. The individual changes in resting pulmonary hemodynamics measured after LVRS were not predictable since they did not correlate with alterations in pulmonary function tests or parameters of gas exchange. Also no differences in the pulmonary artery pressures were found after surgery within the three different emphysema morphology groups (homogeneous, intermediately or markedly heterogeneous distribution of emphysema) [20]. Even in patients with completely homogenous emphysema the PAP mean remained within normal limits. These findings are consistent with the results of Kessler et al. [8], who also found, that resting pulmonary hemodynamics remained unchanged after this type of surgery. Since the cardiac output was not measured, the pulmonary vascular resistance was unable to be calculated. This might be considered a drawback of this study. A more comprehensive investigation of the pulmonary circulation in emphysema would also include measurements of pressures and cardiac output while the patient is breathing air or oxygen (taking measurements at rest and during mild exercise).
Several groups observed a mild rise in the mean PaO 2 [1,2,5,21,22] and a stable or slight decrease in PaCO 2 [1,5,21,22] after bilateral LVRS. In the first 20 patients described by Cooper [1], 10 of 14 did no longer require supplemental oxygen after surgery. In another 53 patients [21], there was a slight but not statistically significant increase in oxygen tension from 62 to 70 mmHg and a decrease in carbon dioxide tension from 43 to 40 mmHg. In the patients a trend towards a slightly higher PaO 2 was found and a mild decrease of PaCO 2 at 6 months after the LVRS. Since the diffusing capacity for carbon monoxide and the alveolo -arterial oxygen partial pressure gradient remained unchanged, it was speculated, that no significant amount of functionally intact gas exchange surface is removed by this procedure. The different effects of LVRS found on blood gases are most likely due to the fact, that emphysema morphology (i.e. very heterogeneous versus more homogeneous type of emphysema, amount of compression of tissue) varies between the various groups, This impedes a fair comparison between the postoperative results in this regard.
In summary, the findings of unaltered resting pulmonary artery pressures and unchanged gas exchange parameters after LVRS are consistent with the speculation, that by this type of surgery mainly functionally useless parts of the lungs are removed, or that the excision of lung pieces with some gas exchange capacity is offset by recruiting of adjacent compressed and potentially less impaired tissue.