Impact of Pulmonary Hypertension on Surgical Mortality and 3-Year Survival After Aortic Valve Replacement

Fundamentos: A hipertensão pulmonar (HP) grave contribui para significante morbidade e mortalidade, e pode aumentar o número de complicações e morte na cirurgia cardíaca. Objetivo: Avaliar o impacto da HP na mortalidade cirúrgica e na sobrevida em três anos, após cirurgia de troca valvar aórtica. Métodos: Estudo de coorte retrospectiva realizado com pacientes submetidos à cirurgia de troca valvar aórtica, de maio de 2011 a dezembro de 2012, no Instituto Nacional de Cardiologia, RJ. A população estudada foi estratificada em dois grupos, de acordo com a pressão sistólica da artéria pulmonar (PSAP) em: <45 mmHg e ≥45 mmHg. Foi utilizado o teste do qui-quadrado para a comparação entre os grupos; a mortalidade cirúrgica foi estudada por regressão logística e a sobrevida pela curva de Kaplan-Meier e teste de log-rank; e a razão de risco pelo método de Cox. Resultados: Estudados 182 pacientes, 57,0% do sexo masculino, idade 61,0±13,0 anos, com hipertensão arterial sistêmica (72,0%), diabetes mellitus (22,0%), classe funcional pela New York Heart Association III/IV (61,1%), PSAP ≥45 mmHg (14,3%). A mortalidade geral foi 12,0%. A sobrevida foi maior no grupo com PSAP <45 mmHg – 89,0% em um ano e 86,0% em três anos, comparado com o grupo com PSAP ≥45 mmHg – 68,8% em um ano e 58,1% (35,1% a 75,4%) em três anos (p=0,0004). Conclusão: Pacientes com PSAP ≥45 mmHg apresentaram maior mortalidade cirúrgica e pior sobrevida em três anos após cirurgia de troca valvar aórtica.


Introduction
Pulmonary hypertension (PH) secondary to left heart disease corresponds to group 2 in the Dana Point 1 classification and this group is currently considered the most prevalent one.The last update of this classification considers the following diseases: left ventricular systolic and diastolic dysfunction, valve diseases and congenital heart diseases 2 .The PH pathophysiology in this context is complex and it is believed that increased left atrial pressure is retrogressively transmitted to the pulmonary vessels to trigger adaptation processes of endothelial cells 3 .
In valve diseases, the prevalence of PH is related to the severity of valve involvement and associated symptoms and is present in about 65% of patients with severe aortic stenosis 4 .When present, PH adds greater morbidity and mortality to these individuals 5,6 .Besides this, it is considered an important prognostic variable in patients with valvulopathy expecting surgery and is associated with complications and death 7 .
In Brazil, despite the limited data about cardiac surgery in the public health system (Brazilian National Health System -SUS), rheumatic disease is responsible for most indications of valve replacement, 8,9 and for the large number of individuals with chronic rheumatic disease receiving clinical treatment 10 .In fact, a recent publication showed that the most common etiologies for valve dysfunction in 491 patients undergoing heart valve surgery were rheumatic disease (60.3%), degenerative valve disease (15.3%) and e n d o c a r d i t i s ( 4 .5 % ) 1 1 .Pulmonary hypertension is part of the delayed progression of valve disease and, in developing countries, there are no studies on the impact of PH on survival, even after the valve replacement surgery.In patients with aortic valve disease, knowledge of data can have great relevance in clinical practice, since recent guidelines do not consider PH for surgery indication 12,13 .Therefore, the purpose of this study was to analyze the impact of PH on surgical mortality and three-year survival after aortic valve replacement surgery.

Methods
This is a cohort retrospective study held at the INC, which included a review of the records of medical records of patients consecutively admitted between May 2011 and December 2012, undergoing cardiac aortic valve replacement surgery, indicated according to the national 14 and international 15 guidelines in the period.
This study was approved by the Research Ethics Committee of Instituto Nacional de Cardiologia (INC) under no.943.667 and was conducted according to the Helsinki Declaration.
The patients were stratified into two groups according to the levels of systolic pulmonary artery pressure (SPAP): <45 mmHg and ≥45 mmHg.Doppler echocardiography device Vivid S6 (GE Healthcare, USA) was used to measure SPAP using the Bernoulli equation.Individuals aged <18, those who had received percutaneous aortic valve implantation and those who underwent aortic valve surgery or double valve replacement surgery (mitral and aortic) were excluded from the study.
Although the surgeries may have varied among surgeons, all procedures were performed using median sternotomy, cardiopulmonary bypass, hypothermia (30-34 °C) and myocardial protection with cardioplegic solution.The type of valve prosthesis was chosen by the assistant team according to the patient's profile or wishes.The Department of Orovalvar Diseases strongly recommends the use of biological prostheses to those older than 65, women with pregnancy plans, and to patients with socioeconomic or educational level that makes it difficult to maintain regular monitoring of anticoagulation.
The INC is a tertiary care hospital with high level of demand for specialized clinical services and highly complex procedures.Patients are regularly monitored (on average, every four months) in an outpatient unit after hospital discharge.Mortality after discharge from index-surgery was assessed by reviewing hospital records and/or by phone calls.The primary outcome studied was post-surgical death, defined as that occurring within 30 days after surgery or during hospitalization related to the index-surgery and three-year survival.Secondary outcomes were length of stay at the hospital and the intensive care unit, time of use of vasoactive drugs and mechanical ventilation after surgery.
Continuous variables were expressed as mean and standard deviation, median and interquartile; and categorical variables were expressed as absolute and relative frequency.Comparisons between continuous variables were performed using the Student t test or the Mann-Whitney U test and one-way analysis of variance (ANOVA) or the Kruskal-Wallis test.To compare categorical variables, chi-square test or Fisher's exact test was used.Odds ratios (OR) were calculated for surgical mortality by logistic regression, following the univariate model and then the multivariate model.Survival functions were calculated employing Kaplan-Meier nonparametric estimator.
Patients were stratified by covariates selected according to their probable prognostic role.The log-rank test was used to compare the survival functions for each covariate.Hazard ratios (HR) were calculated for the prognosis of variables associated with outcomes, with 95% confidence intervals (CI 95%), according to the Cox proportional model.Bivariate Cox analysis was initially performed and followed by the multivariate analysis for factors with a probable role in the outcome (p<0.10).The proportionality of Cox models was assessed using the Schoenfeld residuals diagnosis test.The tests were two-  The number of patients receiving mechanical prostheses was similar between groups.In the group of higher SPAP, there was a higher proportion of patients with tricuspid valvuloplasty, longer mechanical ventilation time and vasoactive drugs (Table 2).
Overall mortality rate was 12.0% (22/182).Post-operative mortality was also higher in the group with higher SPAP (26.9% vs. 9.0%, p=0.008).The presence of SPAP ≥45 mmHg represented an increase in the odds ratio of death of

Discussion
PH is a frequent finding 16 in candidates for aortic valve replacement surgery and may reflect the degree of valve lesion.In this study, 56 (30.0%) patients had SPAP ≥30 mmHg and 14.3% were in the SPAP ≥45 mmHg group.Postoperative mortality was higher in individuals with higher SPAP value with a relative risk of 33.5%.Higher SPAP values were also associated with a greater number of tricuspid valve procedures and a lower proportion of patients requiring amines for less than 24 hours and mechanical ventilation for less than 12 hours.
The moderate/severe PH group also had a lower threeyear survival rate.It is believed that increased SPAP levels are a result of passive transmission of high left ventricular (LV) and left atrial (LA) end-diastolic filling pressures to the pulmonary arteries, indicating the presence of ventricular dysfunction 17 .As expected, patients in the higher SPAP group presented a higher prevalence of NYHA functional class IV and higher systolic and diastolic LV and LA end diameters ratios by SC, as well as lower LV ejection fraction values.Given the high prevalence of rheumatic fever in Brazil, rheumatic etiology contributed to 1/4 of cases of valve replacement.
High levels of SPAP are a known risk factor for morbidity and mortality in cardiac surgery 18 .However, in the context of aortic valve replacement, the impact PH still seems controversial.Johnson et al. 19 found no significant difference in hospital mortality and follow-up of patients with aortic stenosis stratified by SPAP.Tracy et al. 20 also found no difference in mortality in a small sample stratified by SPAP.
Recent retrospective cohort showed that preoperative PH did not affect survival in the follow-up of population and mortality was similar between the mild/moderate and severe SPAP groups; but the study did not compare outcomes between patients with and without PH 21 .In contrast, more recent retrospective cohorts with a larger number of participants have demonstrated the impact of PH not only on surgical mortality but also on patient survival.A study published in 2013 with 1 116 patients 7 found an association between high levels of SPAP and higher surgical mortality and worse survival in mean follow-up time of 2.5 years.Higher SPAP groups also presented higher number of in-hospital complications such as vasoactive amines and mechanical ventilation for longer periods, more infectious complications and longer hospital stay.Melby et al. 22 observed, in 1080 patients with severe aortic stenosis undergoing valve replacement surgery, mortality of 9.4% in those with PH, while in the group without PH this rate was 5.4% (p=0.018).In the same population, the five-year survival rate was significantly lower in groups with higher SPAP.
Analysis of SPAP values after valve replacement surgery showed worse survival for patients with PH (45.0% vs. 78.0%;p<0.001).Miceli et al. 23 and Zuern et al. 24 found, in a sample of 422 and 200 patients, respectively, with severe aortic stenosis, results similar to those mentioned above.In the cohort of Miceli et al. 23 , the five-year survival rate for the normal SPAP group was 86.0%, in contrast to the groups of mild/moderate PH and severe PH, with 81.0% and 63.0%, respectively (p<0.001).
The most important study in a population with aortic regurgitation and PH is probably the one published by Khandhar et al. 25 .Of the 506 patients studied, 16.4% had PH ≥60 mmHg.Of these, 32 underwent aortic valve replacement surgery and a higher five-year survival (62.0% vs. 22.0%; p=0.004) compared with patients treated with medications.Researchers also observed that by subdividing the participants into groups with valve replacement with and without PH and without valve replacement with and without PH, survival was higher in the group with valve replacement and without PH, followed by those with valve replacement and PH, without valve replacement and without PH; and finally, without valve replacement and with PH (p<0.0001).This indicates that even with increased risk of mortality, patients with high SPAP should also be referred for surgery because the outcome is better than just conservative treatment.
Although PH increases morbidity and mortality in cardiac surgery, only four out of 19 risk stratification models consider PH a risk factor for these surgeries 26 .A recent study evaluated the accuracy of the STS risk stratification model for 3 343 patients with PH undergoing the most common cardiac surgeries, of which 24.9% were aortic valve replacement surgery.Researchers found that mortality was significantly higher than predicted by the STS score for moderate and severe PH patients and suggested the inclusion of the variable PH in the risk score model studied 27 .
The current American 11 and European 12 guidelines for the management of valvular heart disease use the presence of PH to decide the moment of intervention in patients with mitral valve disease, but the same does not happen in the aortic valve disease.However, as found in this study and in other studied presented here, it is believed that PH should be considered when discussing the best time for surgery in patients with aortic valve disease, knowing that, when present, moderate to severe PH has a major impact on the survival of patients.

Conclusion
The patients who underwent aortic valve replacement surgery with moderate to severe PH had higher surgical mortality and worse three-year survival.

Potential Conflicts of Interest
This study has no relevant conflicts of interest.

Sources of Funding
This study had no external funding sources.

Academic Association
This study is not associated with any graduate programs.

Table 1 Demographic and clinical characteristics of the study population
the statistical significance was expressed as p<0.05.Stata 11.2 was used.
(*) values expressed in medians and interquartiles; SAH -systemic arterial hypertension; COPD -chronic obstructive pulmonary disease; CrCl -creatinine clearance by Cockcroft-Gault; NYHA IV -New York Heart Association functional class IV; LA/BSA -left atrial diameter by body surface area; LVes/BSA -left ventricular end-systolic volume by BSA; LVed -left ventricular end-diastolic volume; LV -left ventricle; SD -standard deviation tailed and

Table 2 Intraoperative and postoperative surgical characteristics of the study population
(*) values expressed in medians and interquartiles; AF -atrial fibrillation; CABG -coronary artery bypass grafting; MV -mechanical ventilation; CPB -cardiopulmonary bypass; ICU -intensive care unit; TVR -tricuspid valve replacement

Table 4 Cox multivariate survival analysis in the study population
LA/BSA -left atrial diameter by body surface area; NYHA IV -New York Heart Association functional class IV; SPAP -systolic pulmonary artery pressure