- Email: [email protected]
Pulmonary Function Tests Overestimate Chronic Pulmonary Disease in Patients With Severe Aortic Stenosis
Mitchell J. Magee, MD, Morley A. Herbert, PhD, Karen L. Roper, PhD, Elizabeth Holper, MD, Todd M. Dewey, MD, Tricia Snelus, BA, and Michael J. Mack, MD Medical City Dallas Hospital and Cardiopulmonary Research Science and Technology Institute, Dallas, Texas
Background. Pulmonary dysfunction is an important risk factor for postoperative complications after cardiac surgery, and severe chronic obstructive pulmonary disease (COPD) is considered a relative contraindication to aortic valve replacement. Pulmonary function tests may mistakenly diagnose patients as having COPD, when in fact they have pulmonary dysfunction due to heart failure that potentially will improve with valve replacement. Methods. Between January 2009 and July 2011, 214 consecutive patients underwent pulmonary function testing as part of their preoperative screening. Based on the testing, 143 patients were identified as having COPD (52 mild, 42 moderate, and 49 severe), according to The Society of Thoracic Surgery definition. A total of 71 patients had follow-up tests performed at 6 to 12 months postprocedure. Results. A recent smoking history was present in 55 of 214 (25.7%) patients. Aortic valve replacement was performed in 13.6% (29 of 214) of patients by a conventional
surgical approach, in 39.3% (84 of 214) by a transfemoral approach, and in 47.2% (101 of 214) by a transapical approach. Mortality was not significantly different in patients with COPD (12 of 71, 16.9%) compared with patients without COPD (37 of 143, 25.9%), p [ 0.141. Logistic regression analyses failed to identify preoperative COPD severity category (p [ 0.332) as a predictor for mortality. Comparison of pre- and postprocedure tests revealed that 42% (30 of 71) of patients with COPD showed improvement of one COPD severity category or more, including 40% (12 of 30) of patients in the mild group, 43% (9 of 21) of patients in the moderate group, and 45% (9 of 20) of patients in the severe category. Conclusions. Abnormal pulmonary function improves in a significant number of patients with severe aortic stenosis after valve replacement.
A
The purpose of this study was to determine (1) the prevalence and severity of COPD, as defined by The Society of Thoracic Surgeons (STS) Adult Cardiac Surgery Database risk assessment algorithm, in patients with severe AS evaluated for high-risk AVR in the PARTNER trial; (2) the effect of impaired pulmonary function, measured by PFTs and consequent classification into COPD severity groups, on perioperative mortality; and (3) the number of patients who had respiratory insufficiency due to AS rather than pulmonary disease, falsely diagnosed with COPD, and therefore would improve with AVR.
ortic valve replacement (AVR) provides a survival benefit in patients with severe aortic stenosis (AS), even in older age groups. However, associated comorbidities including chronic obstructive pulmonary disease (COPD) may be cause for denial of AVR, increase procedural mortality and morbidity, decrease survival benefit, and delay diagnosis of significant AS. The advent of transcatheter AVR has increased the number of elderly AS patients with significant comorbidities being evaluated for AVR. We observed that many patients screened in our high-risk AVR clinic were diagnosed with COPD based on spirometry, without a history consistent with lung disease, and that some of these patients had improved lung function after AVR. We hypothesized that impaired pulmonary function tests (PFTs) may be due in part to congestive heart failure (CHF) and not COPD.
Accepted for publication April 22, 2013. Presented at the Forty-ninth Annual Meeting of The Society of Thoracic Surgeons, Los Angeles, CA, Jan 26–30, 2013. Address correspondence to Dr Magee, Medical City Dallas Hospital, 7777 Forest Ln, Ste A 307, Dallas, TX 75230; e-mail: mitchell.magee@ hcahealthcare.com.
Ó 2013 by The Society of Thoracic Surgeons Published by Elsevier Inc
(Ann Thorac Surg 2013;96:1329–35) Ó 2013 by The Society of Thoracic Surgeons
Material and Methods Study Population Between January 2009 and July 2011, 214 consecutive patients underwent pulmonary function testing as part of their preoperative screening and enrollment in the PARTNER (Placement of AoRTic traNscathetER) trial utilizing the Edwards Sapien Valve (Edwards Lifesciences, Inc, Irvine, CA). The North Texas Institutional Review Board at Medical City approved the study and all 0003-4975/$36.00 http://dx.doi.org/10.1016/j.athoracsur.2013.04.123
ADULT CARDIAC
Pulmonary Function Tests Overestimate Chronic Pulmonary Disease in Patients With Severe Aortic Stenosis
1330
MAGEE ET AL PFT IN PATIENTS WITH SEVERE AORTIC STENOSIS
Abbreviations and Acronyms
ADULT CARDIAC
AS AVR BNP CABG CHF COPD
= = = = = =
FEV1 PFT STS TA TAVR TF
= = = = = =
aortic stenosis aortic valve replacement b-type natriuretic peptide coronary artery bypass graft congestive heart failure chronic obstructive pulmonary disease forced expiratory volume in 1 second pulmonary function test The Society of Thoracic Surgeons transapical transcatheter aortic valve replacement transfemoral
patients provided informed consent. Patients with critical AS deemed suitable for inclusion in the trial (calculated risk of mortality >10%, aortic valve area of <0.8 cm3) [1] were randomized to undergo conventional, transapical (TA), or transfemoral (TF) AVR. Exclusion criteria have been discussed previously [1]. Patients were grouped according to the STS Adult Cardiac Surgery Database definitions of COPD severity: mild, 60% to 75% of predicted forced expiratory volume in 1 second (FEV1); moderate, 50% to 59% of predicted FEV1; and severe, more than 50% of predicted FEV1. Seventy-one patients with normal PFTs on initial testing were classified as “No COPD” and excluded from further assessment of pulmonary function. Evaluations, including PFTs, were obtained 6 to 12 months postprocedure in the remaining 143 patients when possible. Thirty-seven patients died prior to obtaining follow-up tests, and 35 refused or were unavailable for follow-up PFTs, leaving 71 patients who had PFTs before and after AVR for analysis. Demographic data were extracted from our STS Adult Cardiac Surgery Database and from medical records. Routine blood tests, including b-type natriuretic peptide (BNP) measurements, were obtained preoperatively and 6 to 12 months postoperatively per protocol. Smoking history was obtained from the STS Database, PARTNER Trial case report forms, and patient medical record. In the Surgeon General’s report, studies have shown that the relative risk of death for smokers has decreased to the value of lifelong nonsmokers by 10 to 14 years after smoking cessation [2]. Based on this information, we classified patients who never smoked, as well as smokers who had quit 15 or more years before surgery as nonsmokers. Patients currently smoking, those who had quit within the last 15 years, or those for whom records provided no extra information were classified as current/recent smokers.
Statistical Analysis Summary data were analyzed using SAS 9.3 (SAS Institute, Cary, NC) with categorical variables tested with chisquare or Fisher’s exact tests and continuous variables with t-tests. Paired t-tests were used for repeat measures on individual patients. Survival curves were calculated
Ann Thorac Surg 2013;96:1329–35
using Kaplan-Meier statistics. Logistic regression was carried out with mortality and changes in pulmonary function category as dependent variables, while controlling for a number of independent variables including smoking, sex, and age. A linear regression examined the relationship between changes in measured FEV1 (postoperative – preoperative) and BNP (postoperative – preoperative). The level of statistical significance was set at p < 0.05.
Results Table 1 shows the demographic characteristics of the patients stratified by severity of COPD. No significant differences were identified between groups other than mean age, BNP, smoking status, and pulmonary function. The “No COPD” group had the highest mean age, and a smoking history was more common in the COPD categories compared with no COPD. Mean preoperative BNP was lower in patients without COPD (409.74 pg/mL) compared with those with COPD, but there was not a linear relationship between BNP and COPD severity, with the “Moderate COPD” group having the highest mean BNP level (814.91 pg/mL). Figure 1 shows the study population of patients undergoing preoperative and postoperative pulmonary function testing, grouped according to their COPD severity class as determined from preoperative PFTs. Of the 214 patients, 143 (66.82%) patients with COPD were grouped by severity with 52 (24%) classified as mild, 42 (20%) as moderate, and 49 (23%) as severe. After excluding patients who died and those unavailable for follow-up PFTs, 58% (30 of 52) of the mild group, 50% (21 of 42) of the moderate, and 41% (20 of 49) of the severe groups completed postoperative PFTs and were included in the final analyses. Mortality data were obtained through follow-up according to the trial protocol. Mortality was not significantly different in patients without COPD (12 of 71, 16.9%) compared with those with COPD (37 of 143, 25.9%), p ¼ 0.141. There was no significant difference at last follow-up in survival after AVR, based on preoperative classification of patients as having no COPD (83% alive), or mild (79% alive), moderate (72% alive), or severe COPD (72% alive), p ¼ 0.344. Survival curves for patients stratified by preoperative severity of lung dysfunction are depicted in Figure 2. Univariate and multivariate logistic regression analyses failed to identify any factor predictive of increased mortality, including preoperative COPD severity category (p ¼ 0.332), surgical procedure (p ¼ 0.100), smoking history (p ¼ 0.562), or preoperative BNP levels (p ¼ 0.445). Figure 3 demonstrates the change in COPD severity class when comparing preoperative and postoperative PFTs. After AVR, there was an improvement of one category or more in 40% (12 of 30) of patients in the mild COPD group, 43% (9 of 21) of patients in the moderate group, and 45% (9 of 20) of the severe group. Aggregate pulmonary function measured postoperatively showed significant improvement in all but
Ann Thorac Surg 2013;96:1329–35
Table 1. Comparison of Preoperative Demographic Characteristics of Patients Included in Each of Three Study Groups and the Group Without COPD Variable
Mild COPD (n ¼ 52)
Moderate COPD (n ¼ 42)
Severe COPD (n ¼ 49)
p Value
8.5% (6/71) 53.5% (38/71) 38.0% (27/71) 86.07 6.17 9.44 4.45 409.74 355.35 43.7% (31/71) 53.8% (28/52) 21.2% (11/52) 33.9% (20/59) 11.9% (7/59) 8.5% (5/59) 32.2% (19/59) 25.4% (15/59) 86.4% (51/59) 81.0% (47/58) 0 54.78 11.54 49.2% (29/59) 16.9% (12/71) 2.9% (2/70) 14.1% (10/71) 89.33 16.44 97.41 17.28 117.67 61.65 0.62 0.13 48.18 14.31
15.4% (8/52) 46.2% (24/52) 38.5% (20/52) 82.88 6.45 8.85 4.26 555.76 686.27 38.5% (20/52) 65.0% (26/40) 10.0% (4/40) 43.8% (21/48) 20.8% (10/48) 12.5% (6/48) 29.2% (14/48) 41.7% (20/48) 91.7% (44/48) 87.2% (41/47) 0 54.18 12.79 37.5% (18/48) 19.2% (10/52) 1.9% (1/52) 21.2% (11/52) 66.52 9.13 66.88 6.10 63.85 23.04 0.60 0.13 53.50 18.40
11.9% (5/42) 47.6% (20/42) 40.5% (17/42) 81.79 6.71 10.47 5.07 814.91 688.85 47.6% (20/42) 48.3% (14/29) 20.7% (6/29) 38.2% (13/34) 17.6% (6/34) 11.8% (4/34) 29.4% (10/34) 29.4% (10/34) 91.2% (31/34) 87.9% (29/33) 5.9% (2/34) 51.79 11.98 44.1% (15/34) 23.8% (10/42) 2.4% (1/42) 23.8% (10/42) 55.74 7.98 55.93 6.22 54.69 25.99 0.60 0.10 48.46 19.34
20.4% (10/49) 38.8% (19/49) 40.8% (20/49) 80.49 5.96 9.65 5.08 632.19 648.19 63.3% (31/49) 71.4% (30/42) 9.5% (4/42) 28.9% (13/45) 13.3% (6/45) 2.2% (1/45) 24.4% (11/45) 35.6% (16/45) 86.7% (39/45) 82.9% (34/41) 4.4% (2/45) 52.45 13.38 51.1% (23/45) 46.9% (23/49) 2.0% (1/49) 18.4% (9/49) 49.19 13.63 40.04 6.55 25.29 11.16 0.60 0.12 46.59 14.39
0.285 0.465 0.987 <0.001 0.420 0.006 0.072 0.159 0.265 0.489 0.592 0.297 0.861 0.321 0.773 0.764 0.126 0.559 0.542 0.001 0.321 0.588 <0.001 <0.001 <0.001 0.753 0.170
AoV ¼ aortic valve; AVR ¼ aortic valve replacement; BNP ¼ b-type natriuretic peptide; CABG ¼ coronary artery bypass graft; CHF ¼ congestive heart failure; COPD ¼ chronic obstructive pulmonary disease; FEV1 ¼ forced expiratory volume in 1 second; NYHA ¼ New York Heart Association; pred ¼ predicted; Preop ¼ preoperative; STS ¼ The Society of Thoracic Surgeons; TA ¼ transapical; TAVR ¼ transcatheter aortic valve replacement; TF ¼ transfemoral.
MAGEE ET AL PFT IN PATIENTS WITH SEVERE AORTIC STENOSIS
Conventional AVR TA-type TAVR TF-type TAVR Age, mean SD (years) STS predicted risk mortality (%) Preoperative BNP (pg/mL) Males Previous CABG surgery Previous valve procedure Diabetes Insulin-dependent diabetes Preop renal failure Cerebrovascular disease NYHA - class III/IV CHF Preoperative beta-blocker use Preoperative inotrope use Ejection fraction (%) Peripheral arterial disease Current/recent smoker O2 dependent Mod/severe tricuspid insufficiency Preop forced vital capacity (% pred) Preop FEV1 (% pred) Preop forced expiratory flow 25%–75% AoV area (cm2) Mean aortic gradient (mmHg)
No COPD (n ¼ 71)
1331
ADULT CARDIAC
1332
MAGEE ET AL PFT IN PATIENTS WITH SEVERE AORTIC STENOSIS
Ann Thorac Surg 2013;96:1329–35
ADULT CARDIAC
Fig 1. Study flow showing patients with COPD divided into three groups based on preoperative PFTs who had repeat testing 6 to 12 months after AVR. (AVR ¼ aortic valve replacement; COPD ¼ chronic obstructive pulmonary disease; PFTs ¼ pulmonary functions tests.)
one of the factors measured when compared with preoperative levels (Table 2). There was improvement in postoperative FEV1 compared with preoperative measurements of at least 10% in 42% of the mild COPD category, 50% of the moderate category, and 50% of the severe category patients. A variable representing the change in preoperative COPD severity class after AVR was created, with an improvement of at least one level labeled “Improved,” and either no change or decline of one level or more labeled “Worsened.” Logistic regression analyses were performed to identify predictors of improved function, as defined above, with variables for surgical procedure, smoking status, preoperative COPD class, change in BNP, age, sex, and interaction terms with procedure, smoking status, and preoperative PFT class. Univariate logistic regression analysis only identified change in BNP as a significant predictor of an improved effect (p ¼ 0.031). A decrease in BNP of 100 pg/mL was associated with an improved effect (odds ratio 1.140; 95% confidence interval 1.012 to 1.284).
Fig 2. Survival curves by preoperative COPD severity classification. (COPD ¼ chronic obstructive pulmonary disease; mod ¼ moderate.)
There was also a linear relationship between the change (postoperative – preoperative) in FEV1 and the change in the BNP. Figure 4 demonstrates the linear relationship between decreased BNP and increased FEV1 after AVR. Subset analyses were carried out in the severe COPD group to determine whether surgical approach (conventional, TA, or TF) had an effect on outcomes. After controlling for age and sex differences, no significant effect was noted from the surgical approach (p ¼ 0.106) on mortality or improvement in pulmonary function after AVR (p ¼ 0.184). Similar analyses with the moderate COPD group showed again that surgical approach was not significant for mortality (p ¼ 0.727) or for improved postoperative function (p ¼ 0.669).
Comment Chronic obstructive pulmonary disease is an independent risk factor for developing cardiovascular disease, even after controlling for smoking, and is also a risk factor for
Ann Thorac Surg 2013;96:1329–35
MAGEE ET AL PFT IN PATIENTS WITH SEVERE AORTIC STENOSIS
1333
ADULT CARDIAC
Fig 3. Changes in COPD severity class after surgery compared with preoperative classification. (COPD ¼ chronic obstructive pulmonary disease; mod ¼ moderate; Preop ¼ preoperative.)
cardiovascular-related morbidity and mortality [3]. Patients with COPD are actually more likely to die of cardiovascular complications or cancer than respiratory failure. Chronic obstructive pulmonary disease is also an independent predictor of poor long-term survival after AVR [4, 5]. Chronic obstructive pulmonary disease occurs in 10% or more of the general population aged 75 years or older [6], is as high as 39% in patients with CHF [7], and is diagnosed in approximately 25% of patients with AS [8]. The frequency with which these diagnoses reportedly occur together, as well as numerous symptoms common to both diseases [9–11], highlight the importance of accurately diagnosing each entity when evaluating symptomatic patients at increased risk for both AS and COPD. For use in surgical risk assessment, the STS defines COPD and the severity of disease based on spirometry. Overestimating the severity of COPD in patients with AS may inadvertently deny patients life-saving surgery, while underestimating COPD may result in inaccurate risk assessment of perioperative morbidity and mortality of AVR related to COPD. Abnormal PFTs consistent with the diagnosis of COPD as defined by the STS Adult Cardiac Surgery Database
were common in our study population of high-risk patients undergoing AVR, seen in two thirds of patients. While a significant change in PFTs with AVR would not be expected in patients with underlying chronic lung disease, an improvement would be anticipated in patients with CHF that had been misdiagnosed as COPD. An improvement by at least one COPD severity class was seen in 40% to 45% of patients in each COPD category, demonstrating a high likelihood of misdiagnosis of such patients referred for AVR. This raises the question of whether PFTs alone are the most accurate way to diagnose and risk-stratify such patients. Serum BNP levels are frequently used to diagnose and evaluate the severity of CHF and measure response to treatment. B-type natriuretic peptide levels have also been used to distinguish heart failure from diseases that present with similar clinical signs and symptoms, such as COPD. Baseline BNP measurements may be elevated in patients with COPD compared with patients without COPD, but are not as high as in patients with heart failure. In patients with known COPD, a BNP level of less than 100 pg/mL can be helpful in ruling out significant heart failure, while a BNP level of more than 500 pg/mL can be helpful in ruling in heart failure. While the BNP level was elevated in each group in our study, higher
Table 2. Comparison of Aggregate Preoperative and Postoperative Pulmonary Function Test Results Pulmonary Function Test FEV1 (% predicted) FVC (% predicted) FEF 25%–75% (% predicted) DLCO (% predicted) FEV1/FVC ratio
Preop 56.5 59.6 49.7 59.4 0.96
12.8 12.3 27.3 15.4 0.17
DLCO ¼ diffusion capacity for carbon monoxide; FEF ¼ forced expiratory flow; capacity; Postop ¼ postoperative; Preop ¼ preoperative.
Postop
p Value (Paired t-test)
0.011 0.008 0.006 0.029 0.422
62.1 65.3 62.1 56.8 0.95
20.0 16.5 43.6 20.4 0.17
FEV1 ¼ forced expiratory volume in 1 second;
FVC ¼ forced vital
1334
MAGEE ET AL PFT IN PATIENTS WITH SEVERE AORTIC STENOSIS
Ann Thorac Surg 2013;96:1329–35
ADULT CARDIAC
Fig 4. Correlation between decreased BNP values and improved pulmonary function. (BNP ¼ b-type natriuretic peptide; FEV1 ¼ forced expiratory volume in 1 second; postop ¼ postoperative; preop ¼ preoperative.)
mean levels were seen in the patients with COPD by PFTs. The significant relationship between improved postoperative BNP and FEV1 suggests that BNP should be further studied as an additional marker in surgical patients to assist in distinguishing COPD from CHF. Previous studies have demonstrated COPD as a significant risk factor for postoperative mortality, which was not seen our study. Several possibilities for this finding exist. First, our small sample size may have precluded the ability to detect a significant difference. Second, given the study population of high-risk patients for surgical AVR, multiple other existing comorbidities in all patients may have limited the ability to discern the relative contribution of individual risk factors, or eliminate all relevant interactions between individual variables. The linear correlation between decrease in BNP and improvement in pulmonary function further supports the relationship of preoperative pulmonary dysfunction to heart failure rather than chronic lung disease. Unfortunately, we were unable to identify any specific predictor of mortality or improved pulmonary function after AVR, measurable prior to surgery that might guide clinical decisions regarding surgery or surgical approach. Additional research is warranted to better define preoperatively which patients with abnormal PFTs are more likely to improve with AVR, and patients at increased risk for mortality or major morbidity.
Limitations The patients enrolled in our study included only patients with AS who were at high risk for surgical AVR and thus enrolled in the PARTNER trial. Because COPD may render a patient high risk for AVR, we may have a sample that overestimates this comorbidity. Given the enrollment criteria for PARTNER trial, these data may not be generalizable to all patients undergoing surgical AVR.
Additionally, this was a single-center study at a highvolume transcatheter aortic valve implantation site, again limiting the generalizability of the results. Third, the study population of patients with abnormal PFTs within each COPD category was small, thus it is possible that we may not have been able to detect a true difference in mortality given the sample size. Fourth, there was incomplete follow-up on spirometry on all patients in the study. Lastly, as described by Wang and Petsonk [12], changes observed over time in an individual’s spirometry results could be influenced by a number of technical and measurement factors, seasonal and diurnal variation, as well as shortterm illnesses or chronic alterations in lung health. Currently, however, PFTs are the standard measurement criteria used for such assessment.
References 1. Leon MB, Smith CR, Mack M, et al. Transcatheter aorticvalve implantation for aortic stenosis in patients who cannot undergo surgery. N Engl J Med 2010;363:1597–607. 2. Office on Smoking and Health (US). Health consequences of tobacco use among women. In: Women and Smoking: A Report of the Surgeon General. Atlanta, GA: Centers for Disease Control and Prevention; March 2001:191, http://www. cdc.gov/tobacco/data_statistics/sgr/2001/complete_report/ pdfs/chp3.pdf. Accessed May 29, 2013. 3. Le Jemtel TH, Padeletti M, Jelic S. Diagnostic and therapeutic challenges in patients with coexistent chronic obstructive pulmonary disease and chronic heart failure. J Am Coll Cardiol 2007;49:171–80. 4. Tseng EE, Lee CA, Cameron DE, et al. Aortic valve replacement in the elderly. Risk factors and long-term results. Ann Surg 1997;225:793–802. 5. Kolh P, Kerzmann A, Lahaye L, Gerard P, Limet R. Cardiac surgery in octogenarians; peri-operative outcome and longterm results. Eur Heart J 2001;22:1235–43. 6. Akinbami LJ, Liu X. Chronic obstructive pulmonary disease among adults aged 18 and over in the United States, 1998–2009. NCHS Data Brief 2011;Jun(63):1–8.
7. Mascarenhas J, Azevedo A, Bettencourt P. Coexisting chronic obstructive pulmonary disease and heart failure: implications for treatment, course and mortality. Curr Opin Pulm Med 2010;16:106–11. 8. Faggiano P, Frattini S, Zilioli V, et al. Prevalence of comorbidities and associated cardiac diseases in patients with valve aortic stenosis. Potential implications for the decisionmaking process. Int J Cardiol 2012;159:94–9. 9. Cheng TO. Shortness of breath: COPD or CHF? Int J Cardiol 2005;105:349–50.
MAGEE ET AL PFT IN PATIENTS WITH SEVERE AORTIC STENOSIS
1335
10. Arnold R, Ranchor AV, Koëter GH, de Jongste MJL, Sanderman R. Consequences of chronic obstructive pulmonary disease and chronic heart failure: the relationship between objective and subjective health. Soc Sci Med 2005;61:2144–54. 11. Celli BR, MacNee W, ATS/ERS Task Force. Standards for the diagnosis and treatment of patients with COPD: a summary of the ATS/ERS position paper. Eur Respir J 2004;23:932–46. 12. Wang M, Petsonk EL. Repeated measures of FEV1 over six to twelve months: what change is abnormal? J Occup Environ Med 2004;46:591–5.
DISCUSSION DR GABRIEL ALDEA (Seattle, WA): I am wondering if your conclusion should be qualified a little bit. Despite current definitions, a 50% reduction in FEV1 and DLCO particularly may not really represent severe COPD, as you showed that these correlated with some of the patients’ valve pathology. We are faced now with seeing many such patients, but you only had 2% of your patients on home oxygen, and we are increasingly seeing more patients with FEV1s and DLCOs in the 30% to 40% range. I assume that those patients were excluded, or that many of those were not treated with transcatheter aortic valve replacement (TAVR). Is our definition of severe COPD just too lax and therefore you are not identifying people who really have severe COPD who do not benefit from TAVR? We talked in our institution about whether or not we should do trials of balloon aortic valvuloplasty in some of these people or other more provocative tests to answer that as a bridge to decision making. DR MAGEE: Well, I think you are basically confirming what our supposition was in doing this study in the first place was, are there patients at all severity levels of identified abnormal pulmonary function studies that we are classifying as having significant COPD? And I think you can argue what is significant, whether it is 50%, 40%, 30%, or whether FEV1 is even the right measurement at all to be determining whether or not somebody is going to have increased morbidity and mortality after an aortic valve replacement by any specific approach. Our hope was that we were going to be able to find something, whether it was smoking history, an absolute preop BNP level, or something else that helped us to ferret out which of those patients were really at risk with intrinsic lung disease, but we did not have the numbers to show that. DR ALDEA: We are particularly worried about the patients with pulmonary hypertension and right ventricular dysfunction perhaps as either long-standing CHF or pulmonary dysfunction, but I was wondering what your experience was. DR MAGEE: I agree with you. We actually had right heart catheter data or echo-related Pulmonary vascular resistance information that we were trying to pull together. We did not have enough of that information to include it in our analysis. But that is something we are looking at very critically as well.
DR MICHAEL REARDON (Houston, TX): Well, this is a great, great study as always from the group in Dallas. A couple of things that looked pretty interesting to me that I would like your thoughts on. One is that in every category—none, mild, moderate, and severe—2% were on oxygen. It did not matter how bad your FEV1 was, you were on oxygen at the same rate. And the other one is that although people got better, about half the number got worse in each group, too. So what are you guys going to do going forward? Are you getting more than just FEV1 as you pull this [data] and how is this going to affect your clinical practice? DR MAGEE: Well, we are certainly a lot more circumspect in declining surgery to those patients that have just impairment in their pulmonary function studies and spirometry. I think we are looking at more functional studies whether it is a metabolic heart when feasible, particularly in those patients that fall in what we consider to be a very high-risk category, whether it is 30% or 20% of predicted. And also looking at more functional status measurements whether it is a frailty score or whatever else that you might believe is more effective in determining who might get through an operation with less morbidity and mortality. DR VINOD THOURANI (Atlanta, GA): So in those patients that have severe COPD and a FEV1 of even 25% to 30%, we’ll do a balloon aortic valvuloplasty (BAV). If they get better, then we move forward with some type of therapy. Is that the policy of Dallas or can you just enlighten us a little bit with the severe COPD category patients, does this study change any of that clinical practice as Dr Reardon is talking about? DR MAGEE: Yes. We are being more aggressive in operating on these patients and we have, just as you have, performed a BAV and seen if we have seen some improvement. We have had some patients that, I can think of one patient just within the last 2 months, that I saw who had lung cancer, had marginal PFTs for resection of that lung cancer, went for a TAVR, and came back with much improved lung function, enough to tolerate a lobectomy. So if you time it right, you can offer a lot more to patients that otherwise would not be considered good candidates for anything.
ADULT CARDIAC
Ann Thorac Surg 2013;96:1329–35
Background. Pulmonary dysfunction is an important risk factor for postoperative complications after cardiac surgery, and severe chronic obstructive pulmonary disease (COPD) is considered a relative contraindication to aortic valve replacement. Pulmonary function tests may mistakenly diagnose patients as having COPD, when in fact they have pulmonary dysfunction due to heart failure that potentially will improve with valve replacement. Methods. Between January 2009 and July 2011, 214 consecutive patients underwent pulmonary function testing as part of their preoperative screening. Based on the testing, 143 patients were identified as having COPD (52 mild, 42 moderate, and 49 severe), according to The Society of Thoracic Surgery definition. A total of 71 patients had follow-up tests performed at 6 to 12 months postprocedure. Results. A recent smoking history was present in 55 of 214 (25.7%) patients. Aortic valve replacement was performed in 13.6% (29 of 214) of patients by a conventional
surgical approach, in 39.3% (84 of 214) by a transfemoral approach, and in 47.2% (101 of 214) by a transapical approach. Mortality was not significantly different in patients with COPD (12 of 71, 16.9%) compared with patients without COPD (37 of 143, 25.9%), p [ 0.141. Logistic regression analyses failed to identify preoperative COPD severity category (p [ 0.332) as a predictor for mortality. Comparison of pre- and postprocedure tests revealed that 42% (30 of 71) of patients with COPD showed improvement of one COPD severity category or more, including 40% (12 of 30) of patients in the mild group, 43% (9 of 21) of patients in the moderate group, and 45% (9 of 20) of patients in the severe category. Conclusions. Abnormal pulmonary function improves in a significant number of patients with severe aortic stenosis after valve replacement.
A
The purpose of this study was to determine (1) the prevalence and severity of COPD, as defined by The Society of Thoracic Surgeons (STS) Adult Cardiac Surgery Database risk assessment algorithm, in patients with severe AS evaluated for high-risk AVR in the PARTNER trial; (2) the effect of impaired pulmonary function, measured by PFTs and consequent classification into COPD severity groups, on perioperative mortality; and (3) the number of patients who had respiratory insufficiency due to AS rather than pulmonary disease, falsely diagnosed with COPD, and therefore would improve with AVR.
ortic valve replacement (AVR) provides a survival benefit in patients with severe aortic stenosis (AS), even in older age groups. However, associated comorbidities including chronic obstructive pulmonary disease (COPD) may be cause for denial of AVR, increase procedural mortality and morbidity, decrease survival benefit, and delay diagnosis of significant AS. The advent of transcatheter AVR has increased the number of elderly AS patients with significant comorbidities being evaluated for AVR. We observed that many patients screened in our high-risk AVR clinic were diagnosed with COPD based on spirometry, without a history consistent with lung disease, and that some of these patients had improved lung function after AVR. We hypothesized that impaired pulmonary function tests (PFTs) may be due in part to congestive heart failure (CHF) and not COPD.
Accepted for publication April 22, 2013. Presented at the Forty-ninth Annual Meeting of The Society of Thoracic Surgeons, Los Angeles, CA, Jan 26–30, 2013. Address correspondence to Dr Magee, Medical City Dallas Hospital, 7777 Forest Ln, Ste A 307, Dallas, TX 75230; e-mail: mitchell.magee@ hcahealthcare.com.
Ó 2013 by The Society of Thoracic Surgeons Published by Elsevier Inc
(Ann Thorac Surg 2013;96:1329–35) Ó 2013 by The Society of Thoracic Surgeons
Material and Methods Study Population Between January 2009 and July 2011, 214 consecutive patients underwent pulmonary function testing as part of their preoperative screening and enrollment in the PARTNER (Placement of AoRTic traNscathetER) trial utilizing the Edwards Sapien Valve (Edwards Lifesciences, Inc, Irvine, CA). The North Texas Institutional Review Board at Medical City approved the study and all 0003-4975/$36.00 http://dx.doi.org/10.1016/j.athoracsur.2013.04.123
ADULT CARDIAC
Pulmonary Function Tests Overestimate Chronic Pulmonary Disease in Patients With Severe Aortic Stenosis
1330
MAGEE ET AL PFT IN PATIENTS WITH SEVERE AORTIC STENOSIS
Abbreviations and Acronyms
ADULT CARDIAC
AS AVR BNP CABG CHF COPD
= = = = = =
FEV1 PFT STS TA TAVR TF
= = = = = =
aortic stenosis aortic valve replacement b-type natriuretic peptide coronary artery bypass graft congestive heart failure chronic obstructive pulmonary disease forced expiratory volume in 1 second pulmonary function test The Society of Thoracic Surgeons transapical transcatheter aortic valve replacement transfemoral
patients provided informed consent. Patients with critical AS deemed suitable for inclusion in the trial (calculated risk of mortality >10%, aortic valve area of <0.8 cm3) [1] were randomized to undergo conventional, transapical (TA), or transfemoral (TF) AVR. Exclusion criteria have been discussed previously [1]. Patients were grouped according to the STS Adult Cardiac Surgery Database definitions of COPD severity: mild, 60% to 75% of predicted forced expiratory volume in 1 second (FEV1); moderate, 50% to 59% of predicted FEV1; and severe, more than 50% of predicted FEV1. Seventy-one patients with normal PFTs on initial testing were classified as “No COPD” and excluded from further assessment of pulmonary function. Evaluations, including PFTs, were obtained 6 to 12 months postprocedure in the remaining 143 patients when possible. Thirty-seven patients died prior to obtaining follow-up tests, and 35 refused or were unavailable for follow-up PFTs, leaving 71 patients who had PFTs before and after AVR for analysis. Demographic data were extracted from our STS Adult Cardiac Surgery Database and from medical records. Routine blood tests, including b-type natriuretic peptide (BNP) measurements, were obtained preoperatively and 6 to 12 months postoperatively per protocol. Smoking history was obtained from the STS Database, PARTNER Trial case report forms, and patient medical record. In the Surgeon General’s report, studies have shown that the relative risk of death for smokers has decreased to the value of lifelong nonsmokers by 10 to 14 years after smoking cessation [2]. Based on this information, we classified patients who never smoked, as well as smokers who had quit 15 or more years before surgery as nonsmokers. Patients currently smoking, those who had quit within the last 15 years, or those for whom records provided no extra information were classified as current/recent smokers.
Statistical Analysis Summary data were analyzed using SAS 9.3 (SAS Institute, Cary, NC) with categorical variables tested with chisquare or Fisher’s exact tests and continuous variables with t-tests. Paired t-tests were used for repeat measures on individual patients. Survival curves were calculated
Ann Thorac Surg 2013;96:1329–35
using Kaplan-Meier statistics. Logistic regression was carried out with mortality and changes in pulmonary function category as dependent variables, while controlling for a number of independent variables including smoking, sex, and age. A linear regression examined the relationship between changes in measured FEV1 (postoperative – preoperative) and BNP (postoperative – preoperative). The level of statistical significance was set at p < 0.05.
Results Table 1 shows the demographic characteristics of the patients stratified by severity of COPD. No significant differences were identified between groups other than mean age, BNP, smoking status, and pulmonary function. The “No COPD” group had the highest mean age, and a smoking history was more common in the COPD categories compared with no COPD. Mean preoperative BNP was lower in patients without COPD (409.74 pg/mL) compared with those with COPD, but there was not a linear relationship between BNP and COPD severity, with the “Moderate COPD” group having the highest mean BNP level (814.91 pg/mL). Figure 1 shows the study population of patients undergoing preoperative and postoperative pulmonary function testing, grouped according to their COPD severity class as determined from preoperative PFTs. Of the 214 patients, 143 (66.82%) patients with COPD were grouped by severity with 52 (24%) classified as mild, 42 (20%) as moderate, and 49 (23%) as severe. After excluding patients who died and those unavailable for follow-up PFTs, 58% (30 of 52) of the mild group, 50% (21 of 42) of the moderate, and 41% (20 of 49) of the severe groups completed postoperative PFTs and were included in the final analyses. Mortality data were obtained through follow-up according to the trial protocol. Mortality was not significantly different in patients without COPD (12 of 71, 16.9%) compared with those with COPD (37 of 143, 25.9%), p ¼ 0.141. There was no significant difference at last follow-up in survival after AVR, based on preoperative classification of patients as having no COPD (83% alive), or mild (79% alive), moderate (72% alive), or severe COPD (72% alive), p ¼ 0.344. Survival curves for patients stratified by preoperative severity of lung dysfunction are depicted in Figure 2. Univariate and multivariate logistic regression analyses failed to identify any factor predictive of increased mortality, including preoperative COPD severity category (p ¼ 0.332), surgical procedure (p ¼ 0.100), smoking history (p ¼ 0.562), or preoperative BNP levels (p ¼ 0.445). Figure 3 demonstrates the change in COPD severity class when comparing preoperative and postoperative PFTs. After AVR, there was an improvement of one category or more in 40% (12 of 30) of patients in the mild COPD group, 43% (9 of 21) of patients in the moderate group, and 45% (9 of 20) of the severe group. Aggregate pulmonary function measured postoperatively showed significant improvement in all but
Ann Thorac Surg 2013;96:1329–35
Table 1. Comparison of Preoperative Demographic Characteristics of Patients Included in Each of Three Study Groups and the Group Without COPD Variable
Mild COPD (n ¼ 52)
Moderate COPD (n ¼ 42)
Severe COPD (n ¼ 49)
p Value
8.5% (6/71) 53.5% (38/71) 38.0% (27/71) 86.07 6.17 9.44 4.45 409.74 355.35 43.7% (31/71) 53.8% (28/52) 21.2% (11/52) 33.9% (20/59) 11.9% (7/59) 8.5% (5/59) 32.2% (19/59) 25.4% (15/59) 86.4% (51/59) 81.0% (47/58) 0 54.78 11.54 49.2% (29/59) 16.9% (12/71) 2.9% (2/70) 14.1% (10/71) 89.33 16.44 97.41 17.28 117.67 61.65 0.62 0.13 48.18 14.31
15.4% (8/52) 46.2% (24/52) 38.5% (20/52) 82.88 6.45 8.85 4.26 555.76 686.27 38.5% (20/52) 65.0% (26/40) 10.0% (4/40) 43.8% (21/48) 20.8% (10/48) 12.5% (6/48) 29.2% (14/48) 41.7% (20/48) 91.7% (44/48) 87.2% (41/47) 0 54.18 12.79 37.5% (18/48) 19.2% (10/52) 1.9% (1/52) 21.2% (11/52) 66.52 9.13 66.88 6.10 63.85 23.04 0.60 0.13 53.50 18.40
11.9% (5/42) 47.6% (20/42) 40.5% (17/42) 81.79 6.71 10.47 5.07 814.91 688.85 47.6% (20/42) 48.3% (14/29) 20.7% (6/29) 38.2% (13/34) 17.6% (6/34) 11.8% (4/34) 29.4% (10/34) 29.4% (10/34) 91.2% (31/34) 87.9% (29/33) 5.9% (2/34) 51.79 11.98 44.1% (15/34) 23.8% (10/42) 2.4% (1/42) 23.8% (10/42) 55.74 7.98 55.93 6.22 54.69 25.99 0.60 0.10 48.46 19.34
20.4% (10/49) 38.8% (19/49) 40.8% (20/49) 80.49 5.96 9.65 5.08 632.19 648.19 63.3% (31/49) 71.4% (30/42) 9.5% (4/42) 28.9% (13/45) 13.3% (6/45) 2.2% (1/45) 24.4% (11/45) 35.6% (16/45) 86.7% (39/45) 82.9% (34/41) 4.4% (2/45) 52.45 13.38 51.1% (23/45) 46.9% (23/49) 2.0% (1/49) 18.4% (9/49) 49.19 13.63 40.04 6.55 25.29 11.16 0.60 0.12 46.59 14.39
0.285 0.465 0.987 <0.001 0.420 0.006 0.072 0.159 0.265 0.489 0.592 0.297 0.861 0.321 0.773 0.764 0.126 0.559 0.542 0.001 0.321 0.588 <0.001 <0.001 <0.001 0.753 0.170
AoV ¼ aortic valve; AVR ¼ aortic valve replacement; BNP ¼ b-type natriuretic peptide; CABG ¼ coronary artery bypass graft; CHF ¼ congestive heart failure; COPD ¼ chronic obstructive pulmonary disease; FEV1 ¼ forced expiratory volume in 1 second; NYHA ¼ New York Heart Association; pred ¼ predicted; Preop ¼ preoperative; STS ¼ The Society of Thoracic Surgeons; TA ¼ transapical; TAVR ¼ transcatheter aortic valve replacement; TF ¼ transfemoral.
MAGEE ET AL PFT IN PATIENTS WITH SEVERE AORTIC STENOSIS
Conventional AVR TA-type TAVR TF-type TAVR Age, mean SD (years) STS predicted risk mortality (%) Preoperative BNP (pg/mL) Males Previous CABG surgery Previous valve procedure Diabetes Insulin-dependent diabetes Preop renal failure Cerebrovascular disease NYHA - class III/IV CHF Preoperative beta-blocker use Preoperative inotrope use Ejection fraction (%) Peripheral arterial disease Current/recent smoker O2 dependent Mod/severe tricuspid insufficiency Preop forced vital capacity (% pred) Preop FEV1 (% pred) Preop forced expiratory flow 25%–75% AoV area (cm2) Mean aortic gradient (mmHg)
No COPD (n ¼ 71)
1331
ADULT CARDIAC
1332
MAGEE ET AL PFT IN PATIENTS WITH SEVERE AORTIC STENOSIS
Ann Thorac Surg 2013;96:1329–35
ADULT CARDIAC
Fig 1. Study flow showing patients with COPD divided into three groups based on preoperative PFTs who had repeat testing 6 to 12 months after AVR. (AVR ¼ aortic valve replacement; COPD ¼ chronic obstructive pulmonary disease; PFTs ¼ pulmonary functions tests.)
one of the factors measured when compared with preoperative levels (Table 2). There was improvement in postoperative FEV1 compared with preoperative measurements of at least 10% in 42% of the mild COPD category, 50% of the moderate category, and 50% of the severe category patients. A variable representing the change in preoperative COPD severity class after AVR was created, with an improvement of at least one level labeled “Improved,” and either no change or decline of one level or more labeled “Worsened.” Logistic regression analyses were performed to identify predictors of improved function, as defined above, with variables for surgical procedure, smoking status, preoperative COPD class, change in BNP, age, sex, and interaction terms with procedure, smoking status, and preoperative PFT class. Univariate logistic regression analysis only identified change in BNP as a significant predictor of an improved effect (p ¼ 0.031). A decrease in BNP of 100 pg/mL was associated with an improved effect (odds ratio 1.140; 95% confidence interval 1.012 to 1.284).
Fig 2. Survival curves by preoperative COPD severity classification. (COPD ¼ chronic obstructive pulmonary disease; mod ¼ moderate.)
There was also a linear relationship between the change (postoperative – preoperative) in FEV1 and the change in the BNP. Figure 4 demonstrates the linear relationship between decreased BNP and increased FEV1 after AVR. Subset analyses were carried out in the severe COPD group to determine whether surgical approach (conventional, TA, or TF) had an effect on outcomes. After controlling for age and sex differences, no significant effect was noted from the surgical approach (p ¼ 0.106) on mortality or improvement in pulmonary function after AVR (p ¼ 0.184). Similar analyses with the moderate COPD group showed again that surgical approach was not significant for mortality (p ¼ 0.727) or for improved postoperative function (p ¼ 0.669).
Comment Chronic obstructive pulmonary disease is an independent risk factor for developing cardiovascular disease, even after controlling for smoking, and is also a risk factor for
Ann Thorac Surg 2013;96:1329–35
MAGEE ET AL PFT IN PATIENTS WITH SEVERE AORTIC STENOSIS
1333
ADULT CARDIAC
Fig 3. Changes in COPD severity class after surgery compared with preoperative classification. (COPD ¼ chronic obstructive pulmonary disease; mod ¼ moderate; Preop ¼ preoperative.)
cardiovascular-related morbidity and mortality [3]. Patients with COPD are actually more likely to die of cardiovascular complications or cancer than respiratory failure. Chronic obstructive pulmonary disease is also an independent predictor of poor long-term survival after AVR [4, 5]. Chronic obstructive pulmonary disease occurs in 10% or more of the general population aged 75 years or older [6], is as high as 39% in patients with CHF [7], and is diagnosed in approximately 25% of patients with AS [8]. The frequency with which these diagnoses reportedly occur together, as well as numerous symptoms common to both diseases [9–11], highlight the importance of accurately diagnosing each entity when evaluating symptomatic patients at increased risk for both AS and COPD. For use in surgical risk assessment, the STS defines COPD and the severity of disease based on spirometry. Overestimating the severity of COPD in patients with AS may inadvertently deny patients life-saving surgery, while underestimating COPD may result in inaccurate risk assessment of perioperative morbidity and mortality of AVR related to COPD. Abnormal PFTs consistent with the diagnosis of COPD as defined by the STS Adult Cardiac Surgery Database
were common in our study population of high-risk patients undergoing AVR, seen in two thirds of patients. While a significant change in PFTs with AVR would not be expected in patients with underlying chronic lung disease, an improvement would be anticipated in patients with CHF that had been misdiagnosed as COPD. An improvement by at least one COPD severity class was seen in 40% to 45% of patients in each COPD category, demonstrating a high likelihood of misdiagnosis of such patients referred for AVR. This raises the question of whether PFTs alone are the most accurate way to diagnose and risk-stratify such patients. Serum BNP levels are frequently used to diagnose and evaluate the severity of CHF and measure response to treatment. B-type natriuretic peptide levels have also been used to distinguish heart failure from diseases that present with similar clinical signs and symptoms, such as COPD. Baseline BNP measurements may be elevated in patients with COPD compared with patients without COPD, but are not as high as in patients with heart failure. In patients with known COPD, a BNP level of less than 100 pg/mL can be helpful in ruling out significant heart failure, while a BNP level of more than 500 pg/mL can be helpful in ruling in heart failure. While the BNP level was elevated in each group in our study, higher
Table 2. Comparison of Aggregate Preoperative and Postoperative Pulmonary Function Test Results Pulmonary Function Test FEV1 (% predicted) FVC (% predicted) FEF 25%–75% (% predicted) DLCO (% predicted) FEV1/FVC ratio
Preop 56.5 59.6 49.7 59.4 0.96
12.8 12.3 27.3 15.4 0.17
DLCO ¼ diffusion capacity for carbon monoxide; FEF ¼ forced expiratory flow; capacity; Postop ¼ postoperative; Preop ¼ preoperative.
Postop
p Value (Paired t-test)
0.011 0.008 0.006 0.029 0.422
62.1 65.3 62.1 56.8 0.95
20.0 16.5 43.6 20.4 0.17
FEV1 ¼ forced expiratory volume in 1 second;
FVC ¼ forced vital
1334
MAGEE ET AL PFT IN PATIENTS WITH SEVERE AORTIC STENOSIS
Ann Thorac Surg 2013;96:1329–35
ADULT CARDIAC
Fig 4. Correlation between decreased BNP values and improved pulmonary function. (BNP ¼ b-type natriuretic peptide; FEV1 ¼ forced expiratory volume in 1 second; postop ¼ postoperative; preop ¼ preoperative.)
mean levels were seen in the patients with COPD by PFTs. The significant relationship between improved postoperative BNP and FEV1 suggests that BNP should be further studied as an additional marker in surgical patients to assist in distinguishing COPD from CHF. Previous studies have demonstrated COPD as a significant risk factor for postoperative mortality, which was not seen our study. Several possibilities for this finding exist. First, our small sample size may have precluded the ability to detect a significant difference. Second, given the study population of high-risk patients for surgical AVR, multiple other existing comorbidities in all patients may have limited the ability to discern the relative contribution of individual risk factors, or eliminate all relevant interactions between individual variables. The linear correlation between decrease in BNP and improvement in pulmonary function further supports the relationship of preoperative pulmonary dysfunction to heart failure rather than chronic lung disease. Unfortunately, we were unable to identify any specific predictor of mortality or improved pulmonary function after AVR, measurable prior to surgery that might guide clinical decisions regarding surgery or surgical approach. Additional research is warranted to better define preoperatively which patients with abnormal PFTs are more likely to improve with AVR, and patients at increased risk for mortality or major morbidity.
Limitations The patients enrolled in our study included only patients with AS who were at high risk for surgical AVR and thus enrolled in the PARTNER trial. Because COPD may render a patient high risk for AVR, we may have a sample that overestimates this comorbidity. Given the enrollment criteria for PARTNER trial, these data may not be generalizable to all patients undergoing surgical AVR.
Additionally, this was a single-center study at a highvolume transcatheter aortic valve implantation site, again limiting the generalizability of the results. Third, the study population of patients with abnormal PFTs within each COPD category was small, thus it is possible that we may not have been able to detect a true difference in mortality given the sample size. Fourth, there was incomplete follow-up on spirometry on all patients in the study. Lastly, as described by Wang and Petsonk [12], changes observed over time in an individual’s spirometry results could be influenced by a number of technical and measurement factors, seasonal and diurnal variation, as well as shortterm illnesses or chronic alterations in lung health. Currently, however, PFTs are the standard measurement criteria used for such assessment.
References 1. Leon MB, Smith CR, Mack M, et al. Transcatheter aorticvalve implantation for aortic stenosis in patients who cannot undergo surgery. N Engl J Med 2010;363:1597–607. 2. Office on Smoking and Health (US). Health consequences of tobacco use among women. In: Women and Smoking: A Report of the Surgeon General. Atlanta, GA: Centers for Disease Control and Prevention; March 2001:191, http://www. cdc.gov/tobacco/data_statistics/sgr/2001/complete_report/ pdfs/chp3.pdf. Accessed May 29, 2013. 3. Le Jemtel TH, Padeletti M, Jelic S. Diagnostic and therapeutic challenges in patients with coexistent chronic obstructive pulmonary disease and chronic heart failure. J Am Coll Cardiol 2007;49:171–80. 4. Tseng EE, Lee CA, Cameron DE, et al. Aortic valve replacement in the elderly. Risk factors and long-term results. Ann Surg 1997;225:793–802. 5. Kolh P, Kerzmann A, Lahaye L, Gerard P, Limet R. Cardiac surgery in octogenarians; peri-operative outcome and longterm results. Eur Heart J 2001;22:1235–43. 6. Akinbami LJ, Liu X. Chronic obstructive pulmonary disease among adults aged 18 and over in the United States, 1998–2009. NCHS Data Brief 2011;Jun(63):1–8.
7. Mascarenhas J, Azevedo A, Bettencourt P. Coexisting chronic obstructive pulmonary disease and heart failure: implications for treatment, course and mortality. Curr Opin Pulm Med 2010;16:106–11. 8. Faggiano P, Frattini S, Zilioli V, et al. Prevalence of comorbidities and associated cardiac diseases in patients with valve aortic stenosis. Potential implications for the decisionmaking process. Int J Cardiol 2012;159:94–9. 9. Cheng TO. Shortness of breath: COPD or CHF? Int J Cardiol 2005;105:349–50.
MAGEE ET AL PFT IN PATIENTS WITH SEVERE AORTIC STENOSIS
1335
10. Arnold R, Ranchor AV, Koëter GH, de Jongste MJL, Sanderman R. Consequences of chronic obstructive pulmonary disease and chronic heart failure: the relationship between objective and subjective health. Soc Sci Med 2005;61:2144–54. 11. Celli BR, MacNee W, ATS/ERS Task Force. Standards for the diagnosis and treatment of patients with COPD: a summary of the ATS/ERS position paper. Eur Respir J 2004;23:932–46. 12. Wang M, Petsonk EL. Repeated measures of FEV1 over six to twelve months: what change is abnormal? J Occup Environ Med 2004;46:591–5.
DISCUSSION DR GABRIEL ALDEA (Seattle, WA): I am wondering if your conclusion should be qualified a little bit. Despite current definitions, a 50% reduction in FEV1 and DLCO particularly may not really represent severe COPD, as you showed that these correlated with some of the patients’ valve pathology. We are faced now with seeing many such patients, but you only had 2% of your patients on home oxygen, and we are increasingly seeing more patients with FEV1s and DLCOs in the 30% to 40% range. I assume that those patients were excluded, or that many of those were not treated with transcatheter aortic valve replacement (TAVR). Is our definition of severe COPD just too lax and therefore you are not identifying people who really have severe COPD who do not benefit from TAVR? We talked in our institution about whether or not we should do trials of balloon aortic valvuloplasty in some of these people or other more provocative tests to answer that as a bridge to decision making. DR MAGEE: Well, I think you are basically confirming what our supposition was in doing this study in the first place was, are there patients at all severity levels of identified abnormal pulmonary function studies that we are classifying as having significant COPD? And I think you can argue what is significant, whether it is 50%, 40%, 30%, or whether FEV1 is even the right measurement at all to be determining whether or not somebody is going to have increased morbidity and mortality after an aortic valve replacement by any specific approach. Our hope was that we were going to be able to find something, whether it was smoking history, an absolute preop BNP level, or something else that helped us to ferret out which of those patients were really at risk with intrinsic lung disease, but we did not have the numbers to show that. DR ALDEA: We are particularly worried about the patients with pulmonary hypertension and right ventricular dysfunction perhaps as either long-standing CHF or pulmonary dysfunction, but I was wondering what your experience was. DR MAGEE: I agree with you. We actually had right heart catheter data or echo-related Pulmonary vascular resistance information that we were trying to pull together. We did not have enough of that information to include it in our analysis. But that is something we are looking at very critically as well.
DR MICHAEL REARDON (Houston, TX): Well, this is a great, great study as always from the group in Dallas. A couple of things that looked pretty interesting to me that I would like your thoughts on. One is that in every category—none, mild, moderate, and severe—2% were on oxygen. It did not matter how bad your FEV1 was, you were on oxygen at the same rate. And the other one is that although people got better, about half the number got worse in each group, too. So what are you guys going to do going forward? Are you getting more than just FEV1 as you pull this [data] and how is this going to affect your clinical practice? DR MAGEE: Well, we are certainly a lot more circumspect in declining surgery to those patients that have just impairment in their pulmonary function studies and spirometry. I think we are looking at more functional studies whether it is a metabolic heart when feasible, particularly in those patients that fall in what we consider to be a very high-risk category, whether it is 30% or 20% of predicted. And also looking at more functional status measurements whether it is a frailty score or whatever else that you might believe is more effective in determining who might get through an operation with less morbidity and mortality. DR VINOD THOURANI (Atlanta, GA): So in those patients that have severe COPD and a FEV1 of even 25% to 30%, we’ll do a balloon aortic valvuloplasty (BAV). If they get better, then we move forward with some type of therapy. Is that the policy of Dallas or can you just enlighten us a little bit with the severe COPD category patients, does this study change any of that clinical practice as Dr Reardon is talking about? DR MAGEE: Yes. We are being more aggressive in operating on these patients and we have, just as you have, performed a BAV and seen if we have seen some improvement. We have had some patients that, I can think of one patient just within the last 2 months, that I saw who had lung cancer, had marginal PFTs for resection of that lung cancer, went for a TAVR, and came back with much improved lung function, enough to tolerate a lobectomy. So if you time it right, you can offer a lot more to patients that otherwise would not be considered good candidates for anything.
ADULT CARDIAC
Ann Thorac Surg 2013;96:1329–35