Abstract
Predictors of functional outcomes in patients with chronic thromboembolic pulmonary hypertension (CTEPH) undergoing pulmonary thromboendarterectomy (PTE) are important to identify preoperatively. We hypothesized that baseline severity of pulmonary hypertension and obesity would not be associated with 6-month functional outcomes after PTE. Clinical and hemodynamic data were collected on consecutive patients undergoing PTE from 2008 to 2014. Patients were stratified according to baseline pulmonary vascular resistance (PVR) and body mass index (BMI). Six-minute walk distance (6MWD), New York Heart Association functional class (FC), and echocardiography were assessed in each group at baseline and 6 months after PTE. Regression analyses were performed to evaluate for associations between functional outcomes and baseline PVR and BMI. Forty-two patients underwent PTE and had 6-month follow up data. In comparisons of patients with high and low baseline PVR, the baseline characteristics, distribution of disease, 6MWD, and FC were similar. Postoperative hemodynamics for both groups were similar. At 6 months, both groups achieved improvements in FC, and there were no between-group differences in the change in 6MWD or FC. In comparisons of obese and nonobese patients, perioperative and FC improvement were similar; however, obese patients achieved a greater improvement in 6MWD than nonobese patients (P = 0.04). In conclusion, our data suggest that baseline severity of CTEPH and obesity were not associated with worse functional outcome. Further studies are needed to confirm these results, as these findings could have implications for patient selection for PTE.
Chronic thromboembolic pulmonary hypertension (CTEPH) occurs in 0.1%–4% of patients who suffer from acute pulmonary embolism, and the disease is characterized by incomplete thrombus resolution, leading to pulmonary hypertension (PH) and ultimately right ventricular (RV) dysfunction. 1 CTEPH is unique among the causes of PH in that pulmonary thromboendarterectomy (PTE) is potentially curative in appropriately selected patients; thus, identifying those patients who will benefit from surgery is imperative.
Recent large-scale registry data have reported that patients with more severe PH have higher operative mortality,2,3 but the functional outcome of patients with more severe disease is uncertain. This is likely a reflection of prior publications originating from large referral centers, where patients are seen at the time of surgery and often not followed longitudinally, thus primarily informing on operative outcomes. Given improvements in operative outcomes in more recent reports, 4 and given that patients with severe disease may stand to gain more from surgery, we undertook an evaluation of functional outcome 6 months after PTE, hypothesizing that outcomes in patients with more severe baseline PH would not be different from those in patients with less severe disease.
In addition, we sought to evaluate whether obesity affected outcomes in our population, which is enriched in obese patients. 5 One recent report found no association between baseline body mass index (BMI) and operative mortality; 6 however, few data exist informing on functional outcome in obese patients. Given that numerous studies have observed an “obesity paradox,” where obese patients have a better prognosis after cardiac surgery than nonobese patients, 7 we hypothesized that functional outcomes after PTE in obese patients would not be different from those in nonobese patients.
METHODS
After approval by the Institutional Review Board with waiver of consent (IRB no. 141729), we retrospectively reviewed clinical and hemodynamic data from consecutive patients undergoing PTE between December 2008, when we began performing PTE at our center, and August 2014. Preoperative data were collected, including baseline hemodynamics, pulmonary angiogram, 6-minute walk distance (6MWD), and BMI. CTEPH was diagnosed according to standard criteria, 8 by the presence of at least one perfusion defect on imaging (pulmonary angiogram or ventilation-perfusion lung scan) and a mean pulmonary artery pressure ≥ 25 mmHg on right heart catheterization. Patients selected for surgery met these criteria despite having at least 6 months of anticoagulation treatment. All patients at our center underwent pulmonary angiography unless PTE was recommended urgently after clinical assessment with a confirmatory computed tomography angiogram. Distribution of disease on pulmonary angiogram was retrospectively reviewed by two PH experts (ARH and MEP), who were blinded to patient characteristics, and each case was classified as proximal, segmental, and subsegmental. PTE was performed by a single surgeon (MRP) using standard techniques, as described previously, 9 and perioperative data, including complications and length of stay, were collected. Hemodynamic data were collected immediately after surgery and on the last day measurements were taken during the postoperative period, with reported values being the last hemodynamics recorded with patients not receiving inotropes. Reperfusion pulmonary edema (RPE) was determined as described previously, 10 as worsening hypoxia and radiographic infiltrate in the distribution of endarterectomized pulmonary arteries in the first 72 hours after surgery. Our standard practice is for patients to follow up in our PH clinic within 1 month of surgery and then every 3–6 months. New York Heart Association functional class (FC), echocardiography, 6MWD, and medication use (phosphodiesterase type 5 inhibitors, endothelin receptor antagonists, soluble guanylate cyclase stimulators, and prostanoids) were collected at baseline and at 6 months. The echocardiographic parameters RV function and RV systolic pressure (RVSP) were determined as described previously.11,12
The median pulmonary vascular resistance (PVR) in our patient population was 9.0 Wood units, and we chose to stratify patients into two groups: a low-PVR group (PVR < 9 Wood units) and a high-PVR group (PVR ≥ 9 Wood units). We also analyzed baseline, perioperative, and 6-month data in patients who were obese (BMI ≥ 30) and in nonobese patients (BMI < 30).
Data are expressed as mean ± standard deviation unless otherwise specified. Categorical variables are compared by χ2 analysis and continuous variables by the Mann-Whitney U test. Operative mortality is defined as death within 30 days of surgery. Linear and logistic regressions were also performed to evaluate baseline PVR as a predictor of change in 6MWD and FC, respectively, with adjustments for age and BMI. Linear and logistic regressions were performed to evaluate whether preoperative BMI predicted change in 6MWD and FC, respectively, with adjustments for age and baseline PVR. GraphPad Prism 6 (GraphPad Software, La Jolla, CA) and IBM SPSS Statistics 22 (IBM, Armonk, NY) were used for all analyses.
RESULTS
Sixty-one patients were evaluated and met criteria for CTEPH, and 44 underwent surgery during the study period. The remaining 17 patients did not undergo PTE at our center for the following reasons: 6 were excluded because of medical comorbidities (Table S1; Tables S1–S3 available online), 4 were believed to be inoperable because of predominantly distal disease, 3 patients had normal RV function and minimal symptoms, 2 patients were offered PTE and declined, and 2 patients were referred to the University of California, San Diego, for PTE. Of the 44 patients who underwent surgery, 42 had 6-month follow-up data and were included in the primary analysis. Two patients lived remotely from our center and were lost to follow-up after surgery.
Low- and high-PVR groups
Baseline characteristics for patients with PVRs of <9 and ≥9 Wood units are shown in Tables 1 and S2. Of note, the high- and low-PVR groups were distributed similarly by year during the study period. No major differences in baseline characteristics were observed, including a similar distribution of clot burden, as assessed on pulmonary angiogram, and no differences in CTEPH risk factors, prior venous thromboembolic disease, or the presence of lung disease. On preoperative right heart catheterization, the low-PVR group had evidence of mixed PH with mild elevations in pulmonary artery wedge pressure (PAWP; 14.8 ± 7.3 vs. 9.5 ± 6.4 mmHg, P = 0.03).
Baseline, perioperative, and 6-month outcome data for high- and low-PVR groups
Note: Data are presented as mean ± standard deviation or no. (%). Data were available for all patients except as follows (high-PVR group, low-PVR group): 6MWD (n = 16, 15), RV dysfunction (n = 19, 21), RAP (n = 19, 21), mPAP (n = 19, 21), CI (n = 19, 21), PVR (n = 18, 16), 6-month 6MWD (n = 16, 20), improved 6MWD (n = 14, 15), 6-month RVSP (n = 12, 17), change in RVSP (n = 11, 14), and 6-month RV dysfunction (n = 16, 19). PVR: pulmonary vascular resistance; WU: Wood units; BMI: body mass index; 6MWD: 6-minute walk distance; FC: New York Heart Association functional class; PAH: pulmonary arterial hypertension; RV: right ventricular; RAP: right atrial pressure; mPAP: mean pulmonary artery pressure; PAWP: pulmonary artery wedge pressure; CI: cardiac index (Fick); RVSP: RV systolic pressure.
Perioperative data for both groups are shown in Table 1 with additional data in Table S3. Notably, patients in the high-PVR group, compared to the low-PVR group, had an increased incidence of RPE (45% vs. 14%, P = 0.025). Otherwise, the incidence of individual complications was statistically similar between the two groups, although there were numerically more complications in the high-PVR group. Both groups had significant reductions in PVR; however, the high-PVR group had a greater percentage of reduction in PVR than the low-PVR group (−84.3% ± 10.8% vs. −70.3% ± 18.1%, P = 0.006; Fig. 1). The mean postoperative PVR was similar between the high-PVR group and the low-PVR group (1.9 ± 1.0 vs. 1.8 ± 1.1 Wood units, P = 0.36).
Pulmonary vascular resistance (PVR) before and after pulmonary thromboendarterectomy in the high- and low-PVR groups. Change in PVR before and after surgery, P < 0.0001 for both groups. POD#0: postoperative day 0.
At 6 months, 2 patients were deceased, 1 (from the high-PVR group) 3 weeks after surgery from multiorgan failure believed to result from severe reperfusion lung injury; the second (from low-PVR group) died 4 weeks after surgery from complications related to intracranial hemorrhage while receiving anticoagulation treatment. Data from the remaining 40 survivors are shown in Table 1. The median time of follow-up for the high- and low-PVR groups was 432 and 574 days (P = 0.79), respectively. While patients in the high-PVR group had a greater percentage of reduction in RVSP as assessed on echocardiography, no differences were seen in absolute RVSP or RV function, improvement in 6MWD, need for pulmonary arterial hypertension (PAH) therapy, or improvement in FC. Furthermore, no difference between groups was observed in the proportion of patients who improved FC by ≥1 with surgery (Fig. 2). Functional outcomes at 12 months were analyzed in those patients with 12-month follow-up data (high-PVR group: n = 11; low-PVR group: n = 13), and no significant differences were found in terms of FC, 6MWD, or the need for PAH therapy between the groups. Finally, univariable linear and logistic regression analyses revealed that preoperative PVR was not associated with change in 6MWD or FC, respectively. These findings persisted after adjusting for age and BMI.
New York Heart Association functional class before and 6 months after surgery. PVR: pulmonary vascular resistance (in Wood units).
Obese and nonobese groups
To study whether obese patients had worse outcomes after PTE, we analyzed outcomes in patients with BMI < 30 (n = 21) and patients with BMI ≥ 30 (n = 20), including 7 patients with BMI > 40. Baseline, perioperative, and postoperative data are summarized in Table 2. Before surgery, patients in the high-BMI group had significantly higher right atrial pressures (13.1 ± 5.4 vs. 8.5 ± 4.8 mmHg, P = 0.003) and higher PAWP (15.8 ± 7.0 vs. 9.3 ± 6.2 mmHg, P = 0.002); however, there were no differences in baseline cardiac index (CI) or PVR. No differences in perioperative outcomes were observed, and both groups had similar postoperative hemodynamics. At 6 months, both groups experienced significant improvements in FC from baseline. The high-BMI group had a greater improvement in 6MWD than the low-BMI group (122 ± 99 vs. 53 ± 125 m, P = 0.04); however, the proportion of patients who improved their 6MWD by ≥30 m 13 was not significantly different between obese and nonobese patients (85% and 56%, respectively, P = 0.10). Furthermore, no significant change in BMI was observed in either group after surgery. In unadjusted univariable logistic regression, preoperative BMI was not associated with change in FC, and this did not change after adjustment for age or baseline PVR. Univariable linear regression analysis revealed an association between BMI and change in 6MWD (P = 0.051); however, after adjustment for age and baseline PVR, this association did not persist (P = 0.079).
Baseline, perioperative, and 6-month outcome data for obese and nonobese patients
Note: Data are presented as the mean ± standard deviation or no. (%). Data were available for all patients except as follows (low-BMI group, high-BMI group): RAP (n = 21, 19), mPAP (n = 21, 19), CI (n = 21, 19), PVR (n = 17, 17), 6-month 6MWD (n = 20, 18), improved 6MWD (n = 16, 13). BMI: body mass index; 6MWD: 6-minute walk distance; FC: New York Heart Association functional class; RAP: right atrial pressure; mPAP: mean pulmonary artery pressure; PAWP: pulmonary artery wedge pressure; CI: cardiac index (Fick); PVR: pulmonary vascular resistance; PAH: pulmonary arterial hypertension.
DISCUSSION
Our study examined functional outcomes after PTE in the challenging CTEPH population of patients with severe pulmonary vascular disease or obesity. We found that, in our single center, the perioperative mortality was 2.4% and the 6-month mortality was 4.8%. Our findings are consistent with recent reports of operative mortality after PTE, confirming that PTE can be performed with excellent outcomes by experienced surgeons in carefully selected patients.14,15 Furthermore, the primary finding of this study was that preoperative PVR was not associated with functional outcome 6 months after PTE. We further sought to determine whether obesity was associated with worse outcomes after PTE, and we found that obesity was not associated with higher perioperative mortality or morbidity; however, it may be associated with greater improvements in 6MWD than in nonobese patients.
The definition of disease severity in CTEPH has been an area of active controversy. Prior work by Thistlethwaite et al. 3 had suggested that patients with extreme PH, defined by a systolic pulmonary artery pressure > 100 mmHg, had higher operative mortality than with patients with less severe PH, while other publications have defined severity of CTEPH by PVR.2,16 We chose to use PVR, like more recent publications. Patients were stratified on the basis of the median PVR of 9 Wood units in our study, and this was also in line with the mean or median PVR in recent publications.2,14 We supplemented the primary analysis by performing linear and logistic regression, looking for associations between baseline PVR and functional outcomes.
Several important observations can be made from our data. Clinical variables, including FC and baseline 6MWD, were not different between the high- and low-PVR groups. Disease burden, as assessed by pulmonary angiogram, was similar between groups, and while the majority of patients had proximal disease, a significant burden of segmental and subsegmental disease was present. The lack of distinction between the high- and low-PVR groups in these baseline characteristics is likely multifactorial, reflecting the limitations of FC and 6MWD in assessing disease severity and possibly reflecting selection bias in the selection of patients for PTE.
With regard to hemodynamic assessments before surgery, the low-PVR group had a significantly higher PAWP, suggesting a component of mixed PH in this group. However, preoperative echocardiography did not show a difference in the presence of diastolic dysfunction between the two groups. Given these conflicting findings, further study is needed to evaluate this association and the degree to which patients with mixed group 2 and group 4 PH benefit from PTE.
One notable difference in perioperative outcomes was the higher incidence of RPE in the high-PVR group compared to the low-PVR group. RPE is one of the most common complications causing hypoxemia after PTE, and it is thought to result from excess blood flow to previously occluded pulmonary artery segments, with concomitant hypoperfusion to other regions. 9 RPE has been previously associated with severity of PH, 3 and it is likely that such patients have greater redistribution of blood flow after PTE, leading to heightened ventilation perfusion mismatch and hypoxemia.
Immediately after surgery, the high-PVR group had a greater relative improvement in PVR and CI than the low-PVR group; however, no difference was seen in the final mean PVR or mean CI between the groups. These findings suggest that patients with more severe disease have the most to gain hemodynamically from surgery and that both groups of patients can achieve reductions in PVR to normal or near-normal levels immediately after surgery. No significant difference between the two groups was seen in the absolute RVSP or RV function at 6 months, suggesting that the initial hemodynamic benefits persist beyond the perioperative period. These hemodynamic and echocardiography data strengthen our finding that functional outcomes at 6 and 12 months are similar between the high- and low-PVR groups.
Whether obesity is associated with poorer functional outcomes in CTEPH patients after PTE is unknown. Like the recent report by Fernandes et al., 6 we found no differences in complications or length of stay in the hospital between obese CTEPH patients and nonobese patients. Our study extends beyond the postoperative period by demonstrating similar functional outcomes 6 months after surgery between the two groups. Obese patients had greater absolute improvement in 6MWD from baseline, compared with their nonobese counterparts, although the proportion of patients achieving ≥30-m improvement in 6MWD was not statistically different between the two groups. This paradox may reflect incomplete before-and-after 6MWD data limiting our sample size, and further study of functional outcomes among obese and nonobese patients is needed. Regardless, the data suggest that obese patients can safely undergo PTE with functional outcomes that are at least equivalent to those for nonobese patients. That obese patients did not lose weight after surgery suggests that changes in 6MWD cannot be purely explained by obesity when CTEPH is present. Multivariable linear regression confirmed this possible association between baseline BMI and degree of improvement in 6MWD, highlighting the need to further study PTE in the obese population.
Our study has several limitations. This is a retrospective study and is subject to the inherent limitations of such studies, including but not limited to patient selection bias and the inability to control for confounding variables. Given that this is a single-center study with a relatively small sample size, our results may not be generalizable to other centers, and these factors may partly explain why our findings differ from those of larger studies that have shown that preoperative PVR is related to perioperative survival. 2 Follow-up data were limited beyond 6 months, and functional status at 6 months may not reflect longer-term functional outcomes. Long-term follow-up for these patients is ongoing. Finally, given that our center began performing PTE at the beginning of the study period, it is possible that surgical outcomes and patient selection evolved with time, which could have affected our results. Despite these limitations, we believe that our results are significant and suggest that patients with more severe CTEPH and obese patients may safely undergo surgery with good functional outcomes.
In summary, we examined immediate and 6-month outcomes in patients with higher and lower preoperative PVR and found no significant difference in outcomes. These findings persisted in those patients with 12-month follow-up data. A second analysis suggested that obese patients tolerate PTE well and have at least equivalent functional outcomes compared to nonobese patients. These findings are significant, as a better understanding of functional outcomes beyond the perioperative period is important in assessing which patients with CTEPH may benefit from surgery.
Footnotes
ACKNOWLEDGMENTS
These data were presented as a poster presentation at the American Thoracic Society International Conference in May 2015.