Outcomes of Preprocedural Pulmonary Hypertension on All-Cause and Cardiac Mortality in Patients Undergoing Transcatheter Aortic Valve Implantation: A Systematic Review

Patients with symptomatic aortic stenosis (AS) commonly have an associated finding of pulmonary hypertension (PH), and it has been previously shown to have increased morbidity and mortality following surgical aortic valve repair (SAVR) as well as transcatheter aortic valve implantation (TAVI). There are no guidelines stating the cut-off point for PH at which the patient can safely undergo TAVI with benefits outweighing the risks. This is partly due to the lack of uniformity in the PH definition used in various studies. This systematic review sought to study the effect of preprocedural pulmonary hypertension on early and late all-cause and cardiac mortality in patients undergoing TAVI. We performed a systematic review of studies comparing patients with AS undergoing TAVI having PH. The review was undertaken as per Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines. Articles were identified from PubMed, Pubmed Central (PMC), Cochrane, and Medline on January 10, 2022, for literature published until January 10, 2022. MeSH strategy was used on PubMed to search the literature, and filters were applied to search only Observational Studies, randomized controlled trials (RCT), and meta-analysis. A total of 170 unique articles were identified and screened. Of the 33 full-text articles that were reviewed, 18 articles, including duplicates, were excluded. Fifteen articles fulfilled the selection criteria and were included in this review. The study design included two meta-analyses, one randomized control trial, one prospective cohort study, and 11 retrospective cohort studies. The studies involved a total of approximately 30,000 patients. The observational studies in our review were of good to fair quality, the RCT had a low to moderate bias, and the meta-analysis was of moderate quality. Baseline PH and persistence of PH post-TAVI are strongly associated with all-cause and cardiac mortality. Few studies have shown that a decrease in post-TAVI PH carries mortality benefits. Therefore, efforts should be made to identify mechanisms of persistent PH post-TAVI and whether interventions to reduce PH pre-TAVI will have any clinical implications or not by conducting RCT.


Introduction And Background
Patients with symptomatic aortic stenosis (AS) commonly have an associated finding of pulmonary hypertension (PH), and it has been previously shown to have increased morbidity and mortality following surgical aortic valve repair (SAVR) as well as transcatheter aortic valve implantation (TAVI) [1][2][3][4]. TAVI has been a safe and effective line of treatment for AS and is gaining more attention than SAVR. But because PH carries a higher perioperative risk for any procedure, its benefits versus risk in this population warrants investigation.
PH can be further classified into three main types: pre-capillary PH (Pc-PH), isolated post-capillary PH (Ipc-PH), and combined pre-capillary and post-capillary PH (Cpc-PH). Patients with AS having PH can fall into any of the three categories, but Ipc-PH is the most common subtype [5][6][7]. Ipc-PH can be found in up to 75% of patients with symptomatic AS. It is due to left-sided heart disease.
In the past few years, a new concept of AS-related cardiac injury has been introduced, which is thought to be due to the downstream effects on the left ventricle, left atrium, pulmonary vasculature, and ultimately the right side of the heart [8,9]. Ipc-PH and Cpc-PH represent an advanced stage of the cardiopulmonary process, thought to be due to the functional and structural abnormalities of the aortic valve causing injury to the downstream part of the cardiopulmonary system [10]. A minority of the patients with AS may suffer from Pc-PH or, Ipc-PH, which is not related to AS but attributable to other comorbidities such as chronic lung disease and other structural heart diseases [11].
A limited number of studies have analyzed post-TAVI outcomes among different entities of PH [12][13][14]. In addition, the amount of data on TAVI-induced hemodynamic changes in PH among subtypes of PH is scarce [15][16][17][18]. Fewer studies have reported PH as an independent risk factor post-TAVI and a preprocedural risk factor for TAVI. However, the drawback is that these studies lack a uniform definition of PH [6,[18][19][20][21]. Although it is known that PH is a risk factor and an important determinant of surgical risk, it is not incorporated into the commonly used Society of thoracic surgeon score (STS score), and only severe PH with pulmonary artery systolic pressure (PASP > 60 mm Hg) is considered for European System for Cardiac Operative Risk Evaluation (EuroSCORE).
The evidence for short-and long-term mortality associated with PH after TAVI is inconclusive [11,12,15,17,21]. Few studies have shown the mortality benefits of improving PH post-TAVI [15,18]. No guidelines state the cut-off point for PH at which the patients can safely undergo TAVI, with benefits outweighing the risks. Therefore, this systematic review sought to study the effect of preprocedural PH on early and late all-cause and cardiac mortality in patients undergoing TAVI.

Review Methods
This systematic review was conducted according to Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines [22]. AD and DD performed a literature search in several databases, including PubMed, PubMed Central (PMC), ScienceDirect, Cochrane, and Medline, on January 10, 2022, for literature published between January 1, 2015, and January 10, 2022. MeSH strategy was used on PubMed to search the literature, and filters were applied to search only Observational Studies, Randomized Controlled Trials (RCT), and Meta-Analysis. The following MeSH keywords were used ("Aortic Valve Stenosis") AND ("Transcatheter Aortic Valve Replacement") AND ("Hypertension, Pulmonary"). We applied filters to narrow search results to human subjects, English-only, and by study type to exclude conference abstracts, case reports, comments, editorials, essays, letters, reviews, and unpublished studies. Duplicates were removed. Any additional articles identified by other means were added. We included studies with TAVI, defining the mode of diagnosing PH, papers published in the English language, papers relevant to the question, and papers focusing on outcomes of TAVI with a history of PH. We excluded studies discussing SAVR, papers on PH and AS but no mention of TAVI, papers with a study population of less than 100, unpublished Literature, gray Literature, and low-quality studies.

Study Selection and Data Extraction
The studies were selected through a two-step process. In the first phase, two reviewers (AD and DD) independently ran the search criteria in the databases and reviewed the titles and abstracts to determine eligibility. In the second phase, one reviewer (AD) obtained full-text articles from the initial screen and independently determined if articles were to be included. A third reviewer assessed the validity of the reasons for the excluded articles. Data were extracted from the final set of studies by the first reviewer (AD) and verified by the third reviewer. Extracted data included author name(s), publication year, location, study design, setting, population, size, intervention, method of PH assessment, criteria for PH diagnosis, outcome measures, and results.

Data Analysis
We performed a qualitative analysis of the included studies. Given the heterogeneity of the study and patient population, we could not perform a meta-analysis.

Quality Assessment
The quality of observational studies was assessed using the Newcastle-Ottawa scale, a quality evaluation tool for nonrandomized studies that evaluate studies based on selection, comparability, and outcome [23]. Any study scoring 3-4, 1-2, and 2-3 in the selection, comparability, and outcome/exposure domains were considered "good quality." A study scoring 2, 1-2, and 2-3 in the selection in these domains were considered "good quality." A study scoring 2, 1-2, and 2-3 in these domains were of "fair quality." In all other cases, the study was considered "poor quality." We attempted to minimize study selection bias by having predefined inclusion and exclusion criteria, dual review, and documentation of reasons for excluding studies. The quality appraisal of the studies has been described in Table 1  One study was an RCT that had a low to moderate risk for bias. The two meta-analyses were of moderate quality.

Literature Search
A total of 139 unique articles were identified and screened. Of the 33 full-text articles reviewed, 18 articles, including duplicates, were excluded for reasons listed in Figure 1. Fifteen articles fulfilled the selection criteria and were included in this review.

Study Characteristics
Individual study characteristics are listed in Table 2. All the studies assessed this participant for outcome post-TAVI among patients with PH [1,11,12,[15][16][17][18]21,[24][25][26][27][28][29][30]. Half of the studies were conducted in the United States, the other half were from Europe and one study was from China. The study design included two meta-analyses, one randomized control trial, and 11 retrospective cohort observational studies. The studies involved a total of approximately 30,000 patients. The average population age was around 80 years or more. The studies included patients with severe AS undergoing TAVI. PH was diagnosed either with right heart catheterization (RHC) or echocardiography.    Table 3 lists the results from various studies for short-term and long-term cardiac and all-cause mortality.

Association of Pulmonary Hypertension with Outcomes
Out of 15 studies, 12 studies recorded all-cause mortality in the hospital or at 30 days, out of which eight studies said that PH increases the 30 days of all-cause mortality or in-hospital mortality [11,15,18,21,[24][25][26]30]. Eight studies reported 30 days of cardiac mortality, and five of those studies showed significant 30 days of cardiac mortality associated with pulmonary hypertension. For example, Lindman et al. showed that PH was associated with increased one-year all-cause mortality [29]. Tang   Twelve studies studied late all-cause mortality, concluding that PH was an independent risk factor for late all-cause mortality. Some studies also mentioned that certain subgroups were more likely associated with late all-cause mortality, such as Lindman et al. who showed that both moderate and severe PH was associated with increased mortality [29].  [18]. Residual PH post-procedure was associated with increased one year and at maximal followup of 5.9 years in a study by Alushi et al. [15].  [15]. One prospective study demonstrated a decrease in PASP was significant after TAVI and persistent PASP or new onset severe PH defined as PASP > 60 mm Hg post-TAVI is associated with significant long-term mortality [16].

Discussion
PH can be commonly seen in patients with AS. Studies show that the prevalence of PH in patients undergoing TAVI can range from (12.7% to 75%) [24][25][26][27][28][29][30]. But a national estimate of 103,245 TAVI hospitalizations shows 12.7% had a diagnosis of PH [28]. The most common cause of PH in patients with AS is Ipc-PH due to back pressure changes from the left heart to the pulmonary vasculature. In a minority of the cases of AS undergoing TAVI, patients have Pc-PH or cpc-PH [1,12]. Therefore, we did a qualitative analysis of the studies looking for the prognostic implication of PH on patients undergoing TAVI. We found that pre-TAVI PH was significantly associated with short-term and long-term all-cause mortality and cardiovascular mortality. TAVI is associated with a significant decrease in Pre-TAVI PH, we noticed that post-TAVI persistent PH was associated with increased all-cause and cardiovascular mortality. This effect was most noticed in patients with severe PH post-TAVI.

Definition of PH
The prevalence of PH in the studies selected for this systematic review depended on the diagnostic method of PH and the cut-off value set by the authors. RHC is the gold standard for diagnosing PH and the type. PH can also be predicted non-invasively with the help of echocardiography by using PASP and Tricuspid Regurgitant Velocity (TRV) with additional parameters according to European Society of Cardiology (ESC) guidelines [31]. Two studies were meta-analyses which mainly included studies that assessed PH through echocardiographic parameters [11,30]. Mujeeb et al. grouped PH based on pre-existing diagnoses from the national inpatient database [28]. Six studies included in our systematic review used RHC to assess PH with a cut-off value of mean Pulmonary artery Pressure (mPAP) 25 mm Hg [1,12,17,18,27,29]. Five studies used echocardiographic parameters to suspect PH [15,21,[24][25][26]. Out of those, three studies used PASP with different cut-off values to stratify PH. For example, Barbash [15]. In contrast, a couple of them used TRV to suspect PH, using TRV < 2.8 m/s as the cut-off [25,26]. Table 4 shows the definition of PH used by studies in this analysis.

Study Name Decrease in Pulmonary Hypertension Post Valve Implantation Diagnostic Method for Pulmonary Hypertension
Sinning et al. [16] Improvement in PASP was seen in about 83% of the population, persistent PASP > 60 mm Hg or new onset severe PH was associated with significant mortality post-TAVI. There were no significant differences, either after one year or after four years of follow-up, in patients with Pc-PH and those with Cpc-PH. Interestingly, patients with Ipc-PH showed a significant overall survival benefit after 1 and 4 years in patients with reversible PH.

RHC, mPAP > 25 mm hg
In patients with baseline PH (78%), a significant decrease in PH was observed in 49% at discharge and 59% one year after TAVI. 57% Alushi et al. [15] experienced reductions of their baseline PASPs with an overall mean reduction of 12 -10 mm Hg (57.   Previous studies have shown that patients with PH undergoing SAVR and TAVI have high mortality. In our qualitative analysis, we were able to see a similar trend [1][2][3][4].  [29]. The other three studies, which reported 30 days of cardiovascular mortality but did not find it significantly associated with PH in the TAVI population, had fewer patients studied. Therefore, we think they were not able to reach statistical significance. AF was present in almost 50% of the patient population and can be correlated to the development of Ipc-PH due to retrograde changes and which in turn can cause left atrial enlargement.

Thirty Days Mortality
All-cause mortality: Both the meta-analysis included in our qualitative synthesis found increased all-cause mortality post-TAVI. While Kokkinidis et al. just expanded and validated the meta-analysis by Tang et al., they also analyzed subgroups based on the PASP cut-off for short-term all-cause mortality. Still, it did not come to statistical significance [11,25]. The pooled HR only included five studies and was insignificant, but the RR was significant with moderate heterogeneity [11]. Tang [21,26]. Some studies attributed short-term mortality to the development of acute kidney injury (AKI) post-TAVI, and its prevalence was noted to be (~30% to ~50%) in the included studies. It can also be attributed to prolonged Intensive Care Unit/hospital stays. These findings suggest that there is a need for pre-TAVI management of PH.

Mortality After One Year
Cardiovascular mortality: Two meta-analysis in our study and six other observational studies found PH to be significantly associated with late cardiac mortality. Kokkinidis et al. stated that the PASP > 60 subgroups were at higher risk; only six studies had reported late cardiac mortality in their study [11]. Subgroup analysis of the other metanalysis did not show any statistical significance in increased cardiovascular mortality [30]. While Alushi et al. studied that residual PH post-TAVI was associated with late cardiovascular mortality at one year and a maximum follow-up of 5.9 years [15]. Schewel et al. demonstrated that PH had increased mortality in subgroups with Ipc-PH and pre-capillary PH at one year and four years, but Cpc-PH only significantly increased mortality at four years; they attributed it to a smaller subgroup population [18]. Nijenhuis et al. studied outcomes at two years, high probability of PH was strongly associated with cardiovascular mortality [25]. Testa et al. also reiterated the same thing. Still, they also analyzed the change in PH post-TAVI and reported that severe PH at baseline and persistence of the same post-TAVI is associated with higher mortality [24]. O'Sullivan et al. mentioned that Cpc-PH was significantly associated with the outcome, but they only included the patient who had right heart catheterization in their study, and it might have overestimated their study results; therefore, their results might not be generalizable [1]. The RCTs in our review supported the above studies. Still, they did not have the baseline and post-TAVI echocardiographic parameters. Lindman et al. could not comment on the change in PH post-TAVI and how it affects cardiovascular mortality [29]. Overall, studies analyzing cardiovascular mortality are limited; the ones in the review suggested increased cardiovascular mortality with baseline PH, but data on changes in PASP post-TAVI and its effect on cardiovascular mortality are limited.
All-cause mortality: A meta-analysis from China and the US supported the findings of the observational studies included in our analysis [11,30]. The RCT by Lindman et al. and a retrospective study recommended that along with the objective finding of PH, clinical findings were also important in determining the mortality and should be considered while making decisions for doing a TAVI [25,29]. While Kleczynski et al. demonstrated increased all-cause mortality by stratifying PH using TRV only, they demonstrated no effect on Quality of Life in patients with PH; more studies with a larger cohort are needed to confirm the same [25]. Studies by O'Sullivan and Sultan et al. were only able to demonstrate the correlation of high mortality with Cpc-PH [1,12]. One of them states that the Cpc-PH had severe PH at baseline with a diastolic pressure difference (DPD) > 7 mm Hg, which is known to cause advanced pulmonary vascular remodeling [12]. Similar results were also generated by Schewel et al. as they found cpc-PH at four years and Pcp-PH and Icp-PH at both one year and four years to be significant [18]. Both were biased as they were not consecutive patient series and may not be generalizable to the population. Barbash et al. did a post-hoc analysis and found similar results with severe PH at baseline [21]. Three observational studies reiterated the same thing that persistent PH post-TAVI is associated with increased mortality but had a different definition of persistent PH and its correlation; for example, Testa et al. found a correlation with severe PH (PASP > 60) [24], Masri et al. defined moderate persistent PH post-TAVI (PASP > 45) as an association [17] and as Alushi et al. correlated residual PH or no change in PH or worsening of PH post-TAVI as a predictor of mortality [15].
Overall, all the studies conclude that baseline PH and persistent PH post-TAVI are independent risk factors for late all-cause mortality.

Effect of Decrease in PH
A decrease in PASP after the TAVI has been reported in very few studies, and some of them are included in our review here. As we know, PH can be categorized as Pc-PH, Icp-PH, or Cpc-PH. The studies included in our analysis observed significant improvement in PASP in the Ipc-PH and Cpc-PH groups. The underlying mechanism can explain that: Severe AS leads to increased resistance in the left ventricular outflow, which in turn causes hypertrophy of the left ventricle and progressive diastolic dysfunction. This change causes elevated Left ventricular end-diastolic pressure and, in turn, elevated left atrial filling pressure. All of this causes retrograde changes in the pulmonary veins and, in turn, involves pulmonary arteries. It can be hypothesized that initially severe AS can cause Ipc-PH. Still, as it progresses, it causes intrinsic pulmonary artery changes and eventually involves the right ventricle causing Cpc-PH. Therefore, the most improvement is seen in the patient in the Ipc-PH group.
O'Sullivan, in his study, mentioned that Pc-PH had worse outcomes, and they noticed a significant improvement in PASP and right ventricular function in the Ipc-group [1]. However, they did not study the mortality benefits of decreasing PASP. Barbash et al., in their study, could not prove a post-TAVI decrease in PH and its impact on mortality benefits; we think it is related to their definition of a decrease in PASP post-TAVI as they defined improvement in TAVI as a decrease in PASP ≥ 10 mm Hg [21]. Testa [15]. They also mentioned LVEF > 40%, logistic EuroSCORE < 25%, the presence of severe baseline PH, and the absence of moderate to severe TR as an independent factor for reversible PH. The persistence of PH can be explained by irreversible right ventricular (RV) dysfunction secondary to long-standing elevated PASP, which in turn carries higher mortality. Very few studies here could justify that improvement in PH carries mortality benefits. Overall, PH is a risk factor for TAVI, but it's only included in EuroScore scoring systems. Our studies reported EuroScore and STS scores depending upon the researcher to risk-stratify patients before TAVI.
Therefore, the development of irreversible RV dysfunction poses higher mortality. Therefore, interventions should be made to prevent the progression, which can be done by screening and performing TAVI in the early stage of AS or medically managing the pre-capillary component of PH in the pre-TAVI phase. Table  4 describes the effect of Post TAVI reduction on PH.

Limitations
This was a systematic review, with the majority being observational studies. Therefore, our results should be interpreted in context to that and the biases about observational studies as most of them had a retrospective design. There were only two meta-analyses and one RCT. Some of them were not specifically designed to study PH and its outcomes and there is a lot of heterogeneity. We could not get full-text articles for a couple of studies as they were either paid, not accessible or conference abstracts for which full text was not available. The studies were not uniform in their definition of PH and the method of diagnosing PH, which can underestimate or overestimate study results. Some of the studies lacked the proper definition of cardiovascular mortality.

Conclusions
Baseline PH and persistence of PH post-TAVI are strongly associated with all-cause and cardiac mortality.
RCT are needed to study the cardiovascular effects as very few studies reported cardiovascular mortality.
There is a need to form a uniform definition with the standard diagnostic method for PH in patients undergoing TAVI and include it in the scoring system along with clinical parameters. Few studies have shown that a decrease in post-TAVI PH carries mortality benefits. Therefore, efforts should be made to identify mechanisms of persistent PH post-TAVI and whether interventions to reduce PH pre-TAVI will have any clinical implications or not by conducting RCT. 2023

Conflicts of interest:
In compliance with the ICMJE uniform disclosure form, all authors declare the following: Payment/services info: All authors have declared that no financial support was received from any organization for the submitted work. Financial relationships: All authors have declared that they have no financial relationships at present or within the previous three years with any organizations that might have an interest in the submitted work. Other relationships: All authors have declared that there are no other relationships or activities that could appear to have influenced the submitted work.