Effect of Preoperative Pulmonary Hypertension on Outcomes in Patients With Severe Aortic Stenosis Following Surgical Aortic Valve Replacement




Pulmonary hypertension (PH) is prevalent in patients with aortic stenosis (AS); however, previous studies have demonstrated inconsistent results regarding the association of PH with adverse outcomes after aortic valve replacement (AVR). The goal of this study was to evaluate the effects of preoperative PH on outcomes after AVR. We performed a regional prospective cohort study using the Northern New England Cardiovascular Disease Study Group database to identify 1,116 consecutive patients from 2005 to 2010 who underwent AVR ± coronary artery bypass grafting for severe AS with a preoperative assessment of pulmonary pressures by right-sided cardiac catheterization. PH was defined as a mean pulmonary artery pressure of ≥25 mm Hg, with severity based on the pulmonary artery systolic pressure—mild, 35 to 44 mm Hg; moderate, 45 to 59 mm Hg; and severe, ≥60 mm Hg. We found that PH was present in 536 patients (48%). Postoperative acute kidney injury, low-output heart failure, and in-hospital mortality increased with worsening severity of PH. In multivariate logistic regression, severe PH was independently associated with postoperative acute kidney injury (adjusted odds ratio 4.1, 95% confidence interval [CI] 1.7 to 10, p = 0.002) and in-hospital mortality (adjusted odds ratio 6.9, 95% CI 2.5 to 19.1, p <0.001). There was a significant association between PH and decreased 5-year survival (adjusted log-rank p value = 0.006), with severe PH being associated with the poorest survival (adjusted hazard ratio 2.4, 95% CI 1.3 to 4.2, p = 0.003). In conclusion, severe PH in patients with severe AS is associated with increased rates of in-hospital adverse events and decreased 5-year survival after AVR.


In patients undergoing cardiac surgery, the presence of pulmonary hypertension (PH) is increasingly being recognized as an important prognostic factor associated with increased morbidity and mortality. This is particularly important in patients with severe aortic stenosis (AS) who have an increased prevalence of PH, reported to be as high as 47% to 74%. Previous studies have demonstrated a survival benefit of aortic valve replacement (AVR) in patients with severe AS and severe PH compared with medical therapy. However, previous case series have reported inconsistent results regarding the impact of PH on outcomes after AVR. A possible explanation for this discrepancy is the small sample size of many of these older case series. Recently, larger retrospective single-center studies have demonstrated an association between PH in patients with severe AS and decreased survival after AVR. Confirmation of these findings may potentially have significant clinical implications in patients with severe AS given that the current American College of Cardiology/American Heart Association guidelines for the management of patients with valvular heart disease does not include PH when evaluating the risk of AVR or timing of surgery. Therefore, we used our large regional experience to evaluate the effects of PH on outcomes in patients with severe AS undergoing AVR.


Methods


Patients for this study were drawn from the Northern New England Cardiovascular Disease Study Group Cardiac Surgery Registry. The Northern New England Cardiovascular Disease Study Group is a voluntary regional consortium of physicians, nurses, allied health professionals, hospital administrators, and research scientists from all of the hospitals in Maine, New Hampshire, and Vermont that provide open heart surgery and most hospitals that support percutaneous coronary interventions. The goal of the group is to foster continuous improvement in the quality, safety, and effectiveness of care for patients with cardiovascular disease through the analysis of process and outcomes data with timely feedback to the health-care professionals providing these services. All the hospitals providing an open heart surgery in this region contribute data on consecutive cases with validation of procedure numbers and mortality performed every 2 years. The registry collects prespecified data on patient characteristics including co-morbidities, cardiac history, cardiac anatomy and function, procedural indication, priority, process, and in-hospital outcomes (see http://www.nnecdsg.org/ for the CABG-Valve data form). Relevant to this study, the data collection included fields for mean pulmonary artery pressure (mPAP) and pulmonary artery systolic pressure (PASP).


From 2005 to 2010, 2,478 nonemergent adult patients aged >18 years with severe AS, defined as an aortic valve area of <1 cm 2 , underwent AVR with (n = 1,146) or without (n = 1,332) concomitant coronary artery bypass grafting (CABG) surgery. Patients without a prospective preoperative evaluation of pulmonary pressures by right-sided cardiac catheterization (n = 1,264) or who underwent additional valve or aortic surgery during the same procedure (n = 98) were excluded from the analysis. The final data set consisted of 1,116 patients with 2,663 patient-years of follow-up (mean = 2.39 years).


The focus of our study was to examine the relation between PH and outcomes in patients with severe AS undergoing AVR. PH was defined as an mPAP of ≥25 mm Hg. Patients with PH were divided into 3 groups based on their PASP—mild PH, PASP of 35 to 44 mm Hg; moderate PH, PASP of 45 to 59 mm Hg; and severe PH, PASP of ≥60 mm Hg.


In-hospital outcomes examined included return to the operating room for bleeding, intra- or postoperative stroke, mediastinitis or sternal dehiscence, low-output heart failure (≥1 of the following: intra- or postoperative intra-aortic balloon pump, return to cardiopulmonary bypass for ≥10 minutes, and the need for ≥2 inotropes at 48 hours), postoperative acute kidney injury (AKI, defined as an increase in serum creatinine level to >2.0 mg/dl and 2× the most recent preoperative creatinine level or a new requirement for dialysis after surgery). Patients with end-stage renal disease on dialysis therapy before surgery were excluded from this portion of the analysis (n = 32). Long-term survival was obtained by linkage to the Social Security Administration’s Death Master File, US Department of Commerce Technology Administration and was current through December 2010.


Univariate comparisons across categories of PH were performed using chi-square tests for categorical data and t tests or nonparametric tests of trend for continuous data. p Values of <0.05 (2 tailed) were considered statistically significant. Multivariate logistic regression was used to report adjusted odds ratios (ORs) with 95% confidence intervals (CIs) of PH for the in-hospital outcomes of AKI and mortality, using no PH as the referent group. Crude survival curves were estimated using the nonparametric Kaplan-Meier method. The log-rank test was used to compare crude survival by group. Cox proportional hazards regression was used to determine crude and adjusted hazard ratios (HRs) with 95% CI, using no PH as the referent group. The inverse probability weighting method was used to estimate adjusted survival by hypertension group. The following covariates were used to risk adjust multivariate logistic regression models, HRs, and survival curves: age, gender, preoperative left ventricular ejection fraction (LVEF), 3-vessel coronary disease, left main coronary stenosis, preoperative white blood count of >12,000/μl, recent myocardial infarction within 7 days, urgent priority, previous CABG, peripheral vascular disease, diabetes mellitus, preoperative dialysis, chronic obstructive pulmonary disease, New York Heart Association functional classes III and IV, previous stroke, congestive heart failure, preoperative atrial fibrillation, any coronary artery disease (CAD), preoperative creatinine level of ≥1.3 mg/dl, body surface area, and AVR or AVR + CABG procedure.




Results


During the 6-year period studied, 2,478 patients underwent isolated AVR or AVR + CABG for severe AS. Of these, 1,116 patients (48%) had an invasive hemodynamic measurement of pulmonary pressures recorded and were included in our analysis. PH (mPAP of ≥25 mm Hg) was present in 536 patients (48%), with 218 patients (19.5%) having mild PH (PASP of 35 to 44 mm Hg), 209 patients (18.7%) having moderate PH (PASP of 45 to 59 mm Hg), and 109 patients (9.8%) having severe PH (PASP of ≥60 mm Hg). With increasing severity of PH, there was a significantly greater proportion of patients with an elevated transpulmonary gradient (mPAP − pulmonary capillary wedge pressure) of >10 mm Hg: no PH 15.7%, mild PH 43.1%, moderate PH 55.9%, and severe PH 79.6% (p trend <0.001). The baseline demographics of the patients are listed in Table 1 . Compared with patients without PH, patients with PH were older, had more co-morbid conditions, lower LVEF, and more multivessel CAD. They were more likely to have a recent myocardial infarction and to require urgent surgery. Patients with severe PH were more likely to be older men with advanced heart failure, low LVEF, the greatest burden of CAD, and >1/2 required urgent surgery. Given the greater burden of CAD, patients with PH were more likely to undergo AVR + CABG than patients without PH ( Table 2 ). Those with moderate or severe PH had longer clamp times and pump times and were more likely to receive ≥3 units of packed red blood cells. There was no difference in valve type or valve size by category of PH. Postoperative care differed as well in patients with PH compared with those without PH. PH was associated with a greater use of ≥3 units of packed red blood cells, greater use of inotropes at all time periods, longer times to extubation, and longer lengths of stay.



Table 1

Patient and disease characteristics by pulmonary hypertension (PH) group
















































































































































































































































































Variable PH Group p
None (%) Mild (%) Moderate (%) Severe (%)
Number of procedures 580 218 209 109
Age (yrs)
<60 14.3 11.9 8.1 11.0 0.037
60–69 29.5 25.7 23.9 21.1
70–79 35.3 40.8 39.7 35.8
≥80 20.9 21.6 28.2 32.1
Women 39.0 47.3 45.5 31.2 0.014
Body surface area (m 2 )
<1.70 13.8 11.5 15.3 10.1 0.075
1.70–1.99 44.9 40.8 37.3 33.9
≥2.00 41.3 47.7 47.4 56.0
Preoperative hematocrit (%)
<36 39.2 42.4 50.2 50.5 0.263
36–39 30.6 28.6 24.2 23.9
40–42 17.5 17.1 15.5 12.8
≥43 12.7 12.0 10.1 12.8
Preoperative white blood count >12,000 3.3 4.1 5.7 3.7 0.474
Previous CABG 8.8 6.9 6.7 7.3 0.703
Co-morbid disease
Vascular disease 21.9 25.7 30.1 29.4 0.069
Diabetes mellitus 22.1 32.1 39.2 45.0 <0.001
Chronic obstructive pulmonary disease 15.5 22.9 23.0 24.8 0.011
Congestive heart failure 18.1 27.1 44.5 65.1 <0.001
Dialysis or creatinine level of >2 3.0 3.7 4.3 11.0 0.002
New York Heart Association class IV 4.7 8.7 15.8 26.6 <0.001
Ejection fraction (%)
<40 4.6 8.7 13.6 36.4 <0.001
40–49 5.6 10.7 12.6 12.1
50–59 18.8 18.9 18.9 19.2
≥60 71.0 61.7 55.0 32.3
CAD
Left main stenosis ≥50% 9.0 10.1 10.1 15.6 0.215
3-Vessel disease 10.3 11.9 14.9 20.6 0.019
Myocardial infarction within 7 days 2.8 3.7 6.2 9.2 0.008
Priority at surgery
Elective 77.2 67.9 56.5 42.2 <0.001
Urgent 22.8 32.1 43.5 57.8


Table 2

Intraoperative variables by pulmonary hypertension (PH) group




























































































































































































Variable PH Group p
None (%) Mild (%) Moderate (%) Severe (%)
Number of procedures 580 218 209 109
Primary procedure
AVR 57.8 52.3 46.4 45.0 0.008
CABG + AVR 42.2 47.7 53.6 55.1
Prosthetic valve type
Tissue 85.9 89.0 86.1 90.8 0.738
Mechanical 10.2 7.3 9.1 6.4
Unknown 4.0 3.7 4.8 2.8
Valve size (mm)
Twenty-fifth percentile 21 21 21 21 0.336
Fiftieth percentile (median) 23 23 23 23
Seventy-fifth percentile 23 23 23 25
Clamp time (minutes)
Twenty-fifth percentile 65 65 67.5 73.5 0.003
Fiftieth percentile (median) 81 83 89.5 90
Seventy-fifth percentile 108 102 115 116
Pump time (minutes)
Twenty-fifth percentile 89 93 97 100.5 <0.001
Fiftieth percentile (median) 110 113 126 126
Seventy-fifth percentile 145 139 153 159.5
Red blood cells used intraoperatively
No units 76.2 71.6 72.3 65.1 0.001
1 unit 5 11.0 3.4 11.9
2 units 13.3 11.5 13.4 12.8
≥3 units 5.5 6.0 11 10.1


The overall unadjusted rates of in-hospital outcomes ( Table 3 ) were 3.8% death, 1.8% stroke, 5% AKI, 7.2% low-output heart failure, 4% return to operating room for bleeding, and 0.5% mediastinitis. The development of AKI, low-output heart failure, and in-hospital mortality increased with increasing severity of PH. A similar trend was not seen for stroke or return to the operating room for bleeding. After controlling for differences in case mix and severity of illness, severe PH remained independently associated with an increased risk of postoperative AKI (adjusted OR 4.1, 95% CI 1.7 to 10.0, p = 0.002) and in-hospital mortality (adjusted OR 6.9, 95% CI 2.5 to 19.1, p <0.001). There was an insignificant trend toward increased in-hospital mortality in patients with mild PH (adjusted OR 2.1, 95% CI 0.8 to 5.6, p = 0.16) and moderate PH (adjusted OR 1.8, 95% CI 0.6 to 4.9, p = 0.28) but not for AKI. Severe PH was associated with a nonsignificant increased risk of low-output heart failure (adjusted OR 2.04, 95% CI 0.9 to 4.4, p = 0.07).


Dec 5, 2016 | Posted by in CARDIOLOGY | Comments Off on Effect of Preoperative Pulmonary Hypertension on Outcomes in Patients With Severe Aortic Stenosis Following Surgical Aortic Valve Replacement

Full access? Get Clinical Tree

Get Clinical Tree app for offline access