Coagulation-Fibrinolysis System and Postoperative Outcomes of Patients With Chronic Thromboembolic Pulmonary Hypertension

Abstract

Background: The postoperative changes in the coagulation-fibrinolysis system and the association between the system and postoperative course of patients with chronic thromboembolic pulmonary hypertension (CTEPH) who have undergone pulmonary endarterectomy (PEA) remain unclear.

Methods and Results: Between 1986 and 2013, 117 patients (55.1±11.2 years, preoperative mean pulmonary arterial pressure 46.5±10.5 mmHg) underwent PEA, and 15 patients died during the perioperative period. We studied the association between the preoperative coagulation-fibrinolysis markers and surgical outcomes of all patients, and the long-term outcomes of the 102 survivors from the date of PEA. We also investigated the postoperative changes in coagulation-fibrinolysis markers and their association with residual pulmonary hypertension (PH) in 20 consecutive patients. Only an elevated factor VIII level was associated with perioperative death. Thrombomodulin and plasminogen values were significantly increased after PEA. Univariate logistic regression analysis revealed that D-dimer positivity at follow-up was a risk factor for residual PH. Patients with both an elevated fibrinogen level (≥291 mg/dl [median]) and decreased plasminogen activity (<100% [median]) had significantly worse disease-specific survival than the other patients (5-year disease-specific survival: 84.0% vs. 100%, respectively; P=0.0041 [log-rank test]).

Conclusions: Preoperatively high fibrinogen and low plasminogen values in patients with CTEPH are associated with poor long-term postoperative outcome. PEA benefited not only the pulmonary hemodynamics but also the coagulation-fibrinolysis system of patients. (Circ J 2016; 80: 970–979)

Chronic thromboembolic pulmonary hypertension (CTEPH) is characterized by thrombotic obstruction of the pulmonary arteries and progressive pulmonary hypertension (PH). Pulmonary endarterectomy (PEA) offers symptomatic and prognostic improvement and is a curative treatment for patients with CTEPH.^1–3 Pre- and postoperative hemodynamic parameters have been used to predict the outcomes of patients with operative CTEPH,^4–6 but it is doubtful whether hemodynamic parameters alone can predict the outcome of patients with CTEPH, which has the risk of clot formation.

Editorial p 821

Increased coagulation and decreased fibrinolysis are thought to be involved in the progression of CTEPH. Abnormalities in the coagulation-fibrinolysis system, including elevated levels of tissue type plasminogen activator, type 1 plasminogen activator inhibitor (PAI-1), and clotting factor VIII, and decreased thrombomodulin (TM) levels or activity, have been reported in patients with CTEPH.^7–10 Increased plasma TM levels after PEA have also been reported.⁹ However, there have been few studies that have investigated whether the outcome of patients with CTEPH after PEA is related to abnormalities in the coagulation-fibrinolysis system.

We previously reported that high plasma fibrinogen levels and low plasminogen activity were associated with poor long-term outcomes in medically treated patients with CTEPH.¹¹ Patients with both a high fibrinogen level and low plasminogen activity particularly had extremely poor outcomes. Those results suggest that abnormalities of the coagulation-fibrinolysis system may be associated with poor outcomes, even in operative patients. However, the relationship of the coagulation-fibrinolysis system with the postoperative course of patients with operative CTEPH remains unclear.

Therefore, we tested the following hypotheses: (1) abnormalities in the coagulation-fibrinolysis system are associated with early and long-term outcomes of patients with CTEPH after PEA, and (2) PEA has beneficial effects on the coagulation-fibrinolysis system.

Methods

Study Outline

This was a retrospective study of patients with CTEPH who underwent PEA. The following 3 investigations were performed: (1) the relationship between coagulation-fibrinolysis markers and early and long-term outcomes (disease-specific and overall survival) after PEA; (2) changes in coagulation-fibrinolysis markers between the time of diagnosis and the 1-year follow-up after PEA; (3) the relationship between coagulation-fibrinolysis markers and residual PH at the 1-year follow-up after PEA.

Patients

Between 1986 and 2013, 250 patients were diagnosed with CTEPH at Chiba University Hospital. CTEPH was defined as a mean pulmonary arterial pressure (mPAP) ≥25 mmHg with a normal wedge pressure in patients with dyspnea on exertion for longer than 6 months. An additional requirement was segmental or larger defects on lung perfusion scans in patients with normal ventilation scans. Chronic thromboembolic findings were confirmed on pulmonary angiography. Patients were selected to undergo PEA, based on criteria that were slightly modified from those defined by Moser et al.¹² The study criteria were as follows: (1) mean mPAP >30 mmHg, resulting in a calculated pulmonary vascular resistance (PVR) >300 dyne/s/cm⁵ (3.75 Wood units [WU]), even after oral anticoagulant therapy for >6 months; (2) World Health Organization (WHO) functional class ≥II; (3) thrombi accessible to current surgical techniques (located in main, lobar, or segmental arteries); and (4) absence of severe concomitant disease. There were 134 patients who fulfilled these criteria, consented to surgical therapy, and underwent PEA performed by experienced surgeons. The first 16 patients who underwent PEA via lateral thoracotomy and 1 patient with pneumonia at diagnosis were excluded. Finally, a total of 117 patients treated since 1990 were investigated in this study.

All 117 patients underwent assessments of venous blood samples during diagnostic catheterization, and those data were used for evaluating long-term outcomes. A total of 15 patients died during the perioperative period (within 1 month after PEA), and were excluded from the long-term analysis of the surviving patients (disease-specific survival), resulting in 102 patients who were studied to determine disease-specific survival; 90 of these 102 patients underwent right heart catheterization within 3 months after PEA.

All 102 patients received warfarin; 24 of the 102 patients were preoperatively treated with pulmonary vasodilators at the time of diagnosis (epoprostenol sodium, n=1; beraprost, n=17; bosentan, n=7; ambrisentan, n=1; sildenafil, n=6; tadalafil, n=4), and 28 patients were treated with pulmonary vasodilators during the follow-up period because of progression of CTEPH (epoprostenol sodium, n=1; beraprost, n=12; bosentan, n=4; ambrisentan, n=1; sildenafil, n=10; tadalafil, n=4; riociguat, n=2).

For the evaluation of changes between the time of diagnosis and the 1-year follow-up, a retrospective review revealed that 20 of 117 patients had undergone assessments of pulmonary hemodynamic parameters and coagulation-fibrinolysis markers during that period; 5 patients with mPAP >30 mmHg or requiring additional use of pulmonary vasodilators were identified as having residual PH.^13,14 Pulmonary vasodilators included beraprost (n=7), bosentan (n=4), ambrisentan (n=1), sildenafil (n=2) and tadalafil (n=3) at diagnosis, and bosentan (n=1), sildenafil (n=1), and tadalafil (n=1) at the time of 1-year follow-up because of progression of CTEPH.

This study was approved by the Ethics Committee of Chiba University and written informed consent was given by each patient before catheterization.

Surgical Procedure

PEA was performed via median sternotomy with periods of deep hypothermic circulatory arrest, in accordance with the standardized technique described previously.^4,14 An inferior vena cava filter was inserted in 99 of 117 patients preoperatively.

Data Acquisition

Venous blood samples were collected during right heart catheterization at diagnosis or follow-up. Samples were collected in 3.13% trisodium citrate tubes, and plasma was obtained by centrifuging the tubes at 1,880 g for 10 min at 25℃.

Plasma fibrinogen concentration and clotting factor VIII activity were measured by light-scattering photometry. Plasma plasminogen and α2-plasmin inhibitor (α2PI) activities were measured by a synthesized substrate assay. D-dimer concentration was measured by immune nephelometry. Plasma concentrations of TM, tissue plasminogen activator-plasminogen activator inhibitor-1 complex (TPAIC), thrombin-antithrombin complex (TAT), and plasmin-α2PI complex (PIC) were measured by enzyme-linked immunosorbent assays. Plasma PAI-1 concentration was measured by a latex photometric immunoassay.

Excluding 7 patients who did not have fibrinogen or plasminogen measured, patients were divided into 2 groups based on the median values of 110 patients for plasma fibrinogen concentration (median=291 mg/dl) and plasma plasminogen activity (median=100%). Patients with both a high fibrinogen level and a low plasminogen activity were defined as Group A and all others as Group B, as previously reported.¹¹

Pressure measurements included mPAP and pulmonary arterial wedge pressure (PAWP). Cardiac output (CO) and cardiac index (CI) were estimated using the thermodilution method. PVR was calculated as follows: PVR=(mPAP−PAWP)/CO. Following catheterization, pulmonary angiography was performed using Berman catheters to determine operability.

Patients were diagnosed with coagulopathy if they had positive antiphospholipid antibody, deficiencies in protein C/S and antithrombin, and thrombocytosis.

Comparison of Preoperative Parameters Between Perioperative Survivors and Non-Survivors

In all 117 patients, the clinical and coagulation-fibrinolysis parameters of patients who died during the perioperative period (within 1 month after PEA) and those who survived were compared.

Changes in Coagulation-Fibrinolysis Markers Determined Before and After PEA

For 20 patients who had 1-year follow-up data, clinical parameters such as pulmonary hemodynamics and the coagulation-fibrinolysis parameters determined at the time of diagnosis were compared with those determined at follow-up after PEA.

The clinical and coagulation-fibrinolysis parameters determined at the time of diagnosis were compared with those determined within 3 months after PEA in some patients (n=48).

Assessment of Risk for Residual PH

For the same 20 patients who had 1-year follow-up data, the clinical and coagulation-fibrinolysis parameters determined at the time of diagnosis and at the 1-year follow-up were compared between patients with and without residual PH. Candidate risk factors for residual PH were evaluated by univariate logistic regression analysis. Hemodynamic parameters at the time of 1-year follow-up were not evaluated by univariate logistic regression analysis because residual PH was defined according to those parameters.

Long-Term Survival Analysis

Disease-specific survival and overall survival were calculated from the date of PEA. In April 2014, follow-up data were obtained from all 102 patients either by direct contact with patients or with their primary physicians: 89 patients survived to follow-up and 13 patients did not; 6 of the 13 died of CTEPH-associated right heart failure or sudden death and the other 7 were treated as censored cases (asphyxia, n=1; pneumonia, n=1; interstitial pneumonitis, n=1; cerebral hemorrhage, n=1; cerebral infarction, n=1; multiple organ failure, n=1; natural causes, n=1) for the assessment of disease-specific survival of the perioperative survivors. However, for overall survival analysis, the 7 patients were treated as event counts as well as perioperative deaths.

Candidate predictive factors for disease-specific survival were evaluated by univariate Cox proportional hazards analysis and included age, sex, pulmonary hemodynamic parameters at diagnosis and immediately after PEA, use of pulmonary vasodilators, coagulation-fibrinolysis parameters, and Group A (vs. B), combined factors of plasma fibrinogen level ≥291 mg/dl and plasma plasminogen activity <100% (=Group A) vs. the others (Group B) as previously described. The disease-specific survival of the 98 patients who had preoperative data on fibrinogen and plasminogen were compared after stratification of the patients into Groups A and B. In addition, overall survival, which included perioperative death and death from causes other than CTEPH at follow-up, was compared in patients who had preoperative data on fibrinogen and plasminogen (n=110) and were also similarly stratified.

Statistical Analysis

The results are expressed as mean±SD for continuous variables. Unpaired and paired continuous variables were analyzed using the Wilcoxon-Mann-Whitney test and the Wilcoxon signed-rank test, respectively. Categorical data were analyzed using the chi-square test. The Kaplan-Meier method was used to estimate disease-specific survival and overall survival, with the log-rank test used for comparisons. Univariate Cox proportional hazards analysis was used to evaluate prognostic factors. Univariate logistic regression analysis was used to evaluate risk factors for residual PH. A P-value <0.05 was considered significant. Statistical analysis was performed using JMP version 9.0 software (SAS Institute, Cary, NC, USA).

Results

Patients Characteristics at Diagnosis

The baseline characteristics of the 117 patients who underwent PEA are listed in Table 1. The mean age at diagnosis was 55.1±11.2 years. Of the 117 patients, 44 (37.6%) were male and 73 were female. Most patients (76.1%) were classified as WHO functional class ≥III at diagnosis. The mean baseline mPAP, CI, and PVR were 46.5±10.5 mmHg, 2.6±0.7 L/min/m², and 10.9±4.4 WU, respectively.

Table 1. Baseline Characteristics of All Patients at Diagnosis of CTEPH (n=117)

	Mean±SD, % or n	Median
n	117
Age (years)	55.1±11.2
Sex (male %)	37.6
Coagulopathy (%)	44.2
APA positive (%)	27.1
WHO class III or IV (%)	76.1
D-dimer ≥1 μg/ml (%)	37.6
Fibrinogen (mg/dl)	305.4±71.2	291.0
TAT ≥3.0 ng/ml (%)	11.9
Factor VIII (%)	145.4±72.8
TM (FU/ml)	5.7±2.9
Plasminogen (%)	98.3±16.3	100.0
PIC (μg/ml)	0.9±0.4
PAI-1 (ng/ml)	25.5±11.9
α2PI (%)	103.0±15.6
TPAIC (ng/ml)	13.4±7.8
BNP (pg/ml)	230.2±260.3
PaO2 (mmHg)	50.4±20.2
Group A (%)	17.1
mPAP (mmHg)	46.5±10.5
Cardiac index (L/min/m2)	2.6±0.7
PVR (WU)	10.9±4.4

Coagulopathy was defined as APA positivity, deficiencies in protein C/S and antithrombin, and thrombocytosis. Group A consisted of patients with both high fibrinogen (≥median level of 291 mg/dl) and low plasminogen activity (<median activity of 100%) at diagnosis. α2PI, α2-plasmin inhibitor activity; APA, antiphospholipid antibody; BNP, brain natriuretic peptide; CTE(PH), chronic thromboembolic (pulmonary hypertension); Factor VIII, clotting factor VIII; fibrinogen, plasma fibrinogen; mPAP, mean pulmonary artery pressure; PAI-1, plasminogen activator inhibitor type 1; PIC, plasmin-α2 plasmin inhibitor complex; PVR, pulmonary vascular resistance; TAT, thrombin-antithrombin complex; TM, thrombomodulin; TPAIC, tissue plasminogen activator/PAI-1 complex; WHO, World Health Organization; WU, Wood units.

Postoperative Instability of Coagulation-Fibrinolysis Parameters

The coagulation-fibrinolysis parameters of some patients (n=48) were determined within 3 months after PEA and compared with those parameters at diagnosis (Table S1). The positivity rates of D-dimer and TAT, factor VIII activity, and the concentrations of fibrinogen, TM and PIC were significantly increased after PEA.

Relationship Between Preoperative Parameters and Perioperative Death

The preoperative parameters of patients who died during the perioperative period and those who survived are shown in Table 2. Higher values of factor VIII (213.0±80.7 vs. 138.7±69.1%, P=0.0175), mPAP (52.9±10.3 vs. 45.6±10.2 mmHg, P=0.0102), and PVR (14.3±4.6 vs. 10.3±4.1 WU, P=0.0039) were observed in patients who died perioperatively compared with perioperative survivors, respectively.

Table 2. Preoperative Parameters of CTEPH Patients Who Died Perioperatively and Patients Who Survived

	Perioperative period		P value
	Died	Survived
n	15	102
Age (years)	52.3±11.5	55.5±11.2	NS
Sex (male %)	60.0	34.3	NS
Coagulopathy (%)	40.0	44.9	NS
APA positive (%)	21.4	28.0	NS
WHO class III or IV (%)	86.7	74.5	NS
D-dimer ≥1 μg/ml (%)	28.6	39.0	NS
Fibrinogen (mg/dl)	306.3±65.4	305.3±72.3	NS
TAT ≥3.0 ng/ml (%)	0.0	13.8	NS
Factor VIII (%)	213.0±80.7	138.7±69.1	0.0175
TM (FU/ml)	5.0±2.0	5.7±3.0	NS
Plasminogen (%)	98.2±19.6	98.3±15.9	NS
PIC (μg/ml)	0.9±0.3	0.9±0.4	NS
PAI-1 (ng/ml)	33.8±21.4	24.8±10.6	NS
α2PI (%)	100.6±13.5	103.3±15.9	NS
TPAIC (ng/ml)	16.1±10.1	13.1±7.5	NS
BNP (pg/ml)	293.1±291.2	222.4±256.9	NS
PaO2 (mmHg)	54.3±18.5	49.9±20.5	NS
Group A (%)	8.3	18.8	NS
mPAP (mmHg)	52.9±10.3	45.6±10.2	0.0102
Cardiac index (L/min/m2)	2.3±0.7	2.6±0.7	NS
PVR (WU)	14.3±4.6	10.3±4.1	0.0039

NS, not significant. Other abbreviations as in Table 1.

Effects of PEA on the Coagulation-Fibrinolysis System and Other Parameters at Follow-up

The coagulation-fibrinolysis parameters of 20 patients at the time of diagnosis and at the 1-year (12.9±1.6 months) follow-up after PEA are shown in Table 3. The mean age of these patients at diagnosis was 58.1±11.1 years, and there were more female than male patients. The mean baseline values of mPAP, CI, and PVR were 43.1±10.6 mmHg, 2.9±0.7 L/min/m² and 9.2±3.9 WU, respectively. At the 1-year follow-up after PEA, mPAP, PVR and brain natriuretic peptide (BNP) significantly increased and PaO₂ significantly decreased. Regarding fibrinolytic parameters, TM concentration significantly increased from 4.8±3.3 FU/ml at baseline to 12.8±4.6 FU/ml at the 1-year follow-up (P=0.0002), and plasminogen activity also significantly increased from 100.0±15.0% to 105.9±16.1% (P=0.0312). There were no significant changes in the other markers of the coagulation-fibrinolysis system. The percentage of patients in Group A tended to reduce from 20.0% at base line to 5.0% at 1-year follow-up after PEA, but there was no significance.

Table 3. CTEPH Patients Characteristics and Parameters Assessed at Diagnosis and 1-Year Follow-up (n=20)

	At diagnosis	1-year follow-up	P value
Age (years)	58.1±11.1
Sex (male %)	15.0
Coagulopathy (%)	45.0
APA positive (%)	25.0
WHO class III or IV (%)	60.0	15.0	NS
D-dimer ≥1 μg/ml (%)	30.0	15.0	NS
Fibrinogen (mg/dl)	288.8±52.5	285.0±64.0	NS
TAT ≥3.0 ng/ml (%)	25.0	15.8	NS
Factor VIII (%)	105.4±38.0	119.0±43.6	NS
TM (FU/ml)	4.8±3.3	12.8±4.6	0.0002
Plasminogen (%)	100.0±15.0	105.9±16.1	0.0312
PIC (μg/ml)	0.7±0.4	0.7±0.3	NS
PAI-1 (ng/ml)	23.8±7.5	21.2±10.3	NS
α2PI (%)	100.5±12.2	102.2±10.7	NS
TPAIC (ng/ml)	12.6±6.1	14.7±6.3	NS
BNP (pg/ml)	162.8±228.6	62.4±60.3	0.0400
PaO2 (mmHg)	57.0±8.8	71.3±17.7	0.0006
Group A (%)	20.0	5.0	NS
mPAP (mmHg)	43.1±10.6	24.6±8.5	<0.0001
Cardiac index (L/min/m2)	2.9±0.7	2.8±0.6	NS
PVR (WU)	9.2±3.9	3.8±2.2	<0.0001

Abbreviations as in Tables 1,2.

Risk for Residual PH at Follow-up

The same 20 patients were analyzed for risk factors of residual PH at the 1-year follow-up; 5 of the 20 had residual PH and their preoperative and 1-year follow-up data were used for these analyses.

There were no significant differences in coagulation-fibrinolysis parameters between the preoperative parameters of patients with and without residual PH at the 1-year follow-up (Table 4).

Table 4. CTEPH Patients Characteristics and Parameters Assessed at Diagnosis of Patients With and Without Residual PH at 1-Year Follow-up (n=20)

Preoperative data	1-year residual PH		P value
	(−) n=15	(+) n=5
Age (years)	56.1±11.7	64.0±7.0	NS
Sex (male %)	13.3	20.0	NS
Coagulopathy (%)	40.0	60.0	NS
APA positive (%)	20.0	40.0	NS
WHO class III or IV (%)	53.3	80.0	NS
D-dimer ≥1 μg/ml (%)	26.7	40.0	NS
Fibrinogen (mg/dl)	289.9±53.0	285.6±57.1	NS
TAT ≥3.0 ng/ml (%)	26.7	20.0	NS
Factor VIII (%)	105.7±11.1	104.6±28.4	NS
TM (FU/ml)	4.4±2.4	6.0±6.1	NS
Plasminogen (%)	100.5±15.9	98.6±13.2	NS
PIC (μg/ml)	0.7±0.4	0.7±0.1	NS
PAI-1 (ng/ml)	25.2±7.8	19.6±5.0	NS
α2PI (%)	99.4±13.6	103.6±6.8	NS
TPAIC (ng/ml)	13.8±5.9	8.9±5.7	NS
BNP (pg/ml)	106.5±95.0	331.8±411.1	NS
PaO2 (mmHg)	57.5±9.0	55.7±8.7	NS
Group A (%)	20.0	20.0	NS
mPAP (mmHg)	39.5±8.8	54.0±8.0	0.0126
Cardiac index (L/min/m2)	2.8±0.7	3.0±0.7	NS
PVR (WU)	8.2±3.3	12.5±3.8	0.0447

Abbreviations as in Tables 1,2.

Table 5A shows the postoperative parameters at 1-year follow-up of the patients with and without residual PH. Significantly increased values of BNP, mPAP, and PVR, and a higher D-dimer positivity rate were observed in the patients with residual PH. Univariate logistic regression analysis revealed that D-dimer positivity and a high BNP level at follow-up were associated with risk for residual PH (Table 5B).

Table 5. Residual PH at the 1-Year Follow-up (n=20), (A) CTEPH Patients Characteristics and Parameters Assessed at the 1-Year Follow-up of Patients With and Without 1-Year Residual PH, (B) Risk Factors for Residual PH at the 1-Year Follow-up

(A)
1-year follow-up data	1-year residual PH		P value
	(−) n=15	(+) n=5
WHO class III or IV (%)	6.7	40.0	NS
D-dimer ≥1 μg/ml (%)	0.0	60.0	0.0011
Fibrinogen (mg/dl)	285.5±68.9	283.2±53.6	NS
TAT ≥3.0 ng/ml (%)	13.3	25.0	NS
Factor VIII (%)	122.9±42.7	104.3±50.6	NS
TM (FU/ml)	11.9±4.6	16.2±2.9	NS
Plasminogen (%)	104.8±18.3	109.0±6.1	NS
PIC (μg/ml)	0.7±0.3	0.8±0.3	NS
PAI-1 (ng/ml)	20.8±10.8	22.5±9.3	NS
α2PI (%)	100.9±9.6	106.0±14.1	NS
TPAIC (ng/ml)	14.4±6.7	15.8±4.9	NS
BNP (pg/ml)	44.7±39.1	115.6±85.1	0.0402
PaO2 (mmHg)	74.5±14.5	62.5±24.3	NS
Group A (%)	6.7	0.0	NS
mPAP (mmHg)	20.7±4.2	36.4±6.7	0.0012
Cardiac index (L/min/m2)	2.8±0.7	2.8±0.5	NS
PVR (WU)	2.7±0.8	7.2±1.4	0.0012
(B)
1-year follow-up data	Univariate logistic regression analysis		P value
	OR　　　95% CI
WHO class III or IV	9.3333 (0.6850–246.3649)		NS
D-dimer ≥1 μg/ml	125,230,612 (6.0698–　　　)		0.0014
Fibrinogen	0.9994 (0.9816–1.0158)		NS
TAT ≥3.0 ng/ml	2.1667 (0.0844–31.8684)		NS
Factor VIII	0.9900 (0.9624–1.0163)		NS
TM	1.3255 (0.9805–2.0961)		NS
Plasminogen	1.0164 (0.9506–1.0890)		NS
PIC	3.0198 (0.0348–379.2041)		NS
PAI-1	1.0164 (0.9003–1.1297)		NS
α2PI	1.0485 (0.9502–1.1704)		NS
TPAIC	1.0379 (0.8623–1.2594)		NS
BNP	1.0248 (1.0032–1.0612)		0.0202
PaO2	0.9551 (0.8768–1.0181)		NS
Group A	0.0000 (　　　–18.3810)		NS

Continuous variables were indicated as per unit increase. Upper limit of CI for D-dimer could not be accurately calculated by the statistical software because the value was too high. Lower limit of CI for Group A (vs. B) could not be accurately calculated because the value was too low. CI, confidence interval; OR, odds ratio. Other abbreviations as in Tables 1,2.

Disease-Specific and Overall Survival and Prognostic Factors

The baseline characteristics of all 102 perioperative survivors are shown in Table 6. The mean age at diagnosis was 55.5±11.2 years. Of the 102 patients, 35 (34.3%) were male and 67 were female. There were 74.5% of patients classified as WHO functional class ≥III. The mean baseline values of mPAP, CI, and PVR were 45.6±10.2 mmHg, 2.6±0.7 L/min/m², and 10.3±4.1 WU, respectively. Postoperative hemodynamic data were obtained for 85 of 102 patients within 3 months (1.38±0.44 months) after PEA. The mean postoperative values of mPAP, CI, and PVR were 24.6±9.7 mm Hg, 3.1±0.6 L/min/m² and 3.9±2.7 WU, respectively.

Table 6. Characteristics and Parameters Assessed at Diagnosis of CTEPH Patients Included in the Disease-Specific Survival Analysis

Parameter	Mean±SD, % or n	Median
n	102
Data at diagnosis
Age (years)	55.5±11.2
Sex (male %)	34.3
Coagulopathy (%)	44.9
APA positive (%)	28.0
WHO class III or IV (%)	74.5
D-dimer ≥1 μg/ml (%)	38.9
Fibrinogen (mg/dl)	305.3±72.3	291.0
TAT ≥3.0 ng/ml (%)	13.8
Factor VIII (%)	138.7±69.1
TM (FU/ml)	5.7±3.0
Plasminogen (%)	98.3±15.9	100.0
PIC (μg/ml)	0.9±0.4
PAI-1 (ng/ml)	24.8±10.6
α2PI (%)	103.3±15.9
TPAIC (ng/ml)	13.1±7.5
BNP (pg/ml)	222.4±256.9
PaO2 (mmHg)	49.9±20.5
Group A (%)	18.2
mPAP (mmHg)	45.6±10.2
Cardiac index (L/min/m2)	2.6±0.7
PVR (WU)	10.3±4.1
Data at follow-up within 3 months
Postoperative mPAP	24.6±9.7
Postoperative cardiac index	3.1±0.6
Postoperative PVR	3.9±2.7
Over follow-up period
Use of pulmonary vasodilators (%)*	29.8

*Pulmonary vasodilators included epoprostenol sodium, beraprost, bosentan, ambrisentan, sildenafil, tadalafil, and riociguat. Abbreviations as in Tables 1,2.

The mean observation period was 6.7±5.0 years. The 5-year disease-specific survival of all 102 patients who survived PEA was 97.3%. The characteristics of the 98 patients with preoperative data on fibrinogen and plasminogen who were stratified according to median fibrinogen and plasminogen values are shown in Table S2A. Group A (n=17 patients with increased fibrinogen and decreased plasminogen) had higher TPAIC levels than Group B (n=81) and all other patients (15.8±8.3 vs. 12.4±7.3 ng/ml, respectively; P=0.0288). Group A had a significantly worse 5-year disease-specific survival than Group B (84.0% vs. 100%, respectively; log-rank test P=0.0041; Figure A). Univariate Cox proportional hazards analysis revealed that older age, Group A (vs. B), higher postoperative mPAP, and higher postoperative PVR were significant prognostic factors (Table 7).

Figure.

(A) Kaplan-Meier estimates of 5-year disease-specific survival in patients who survived pulmonary endarterectomy. Patients were classified by median plasma fibrinogen level and plasminogen activity. Group A (red), 17 patients with both high fibrinogen and low plasminogen; Group B (blue), the other 81 patients. Group A had significantly worse disease-specific survival than Group B (5-year survival, 84.0% vs. 100%, respectively; P=0.0041, log-rank test). (B) Kaplan-Meier survival estimates of overall 5-year survival of all patients with chronic thromboembolic pulmonary hypertension, including patients who died during the perioperative period (within 1 month after pulmonary endarterectomy) and who died from other causes. Patients were classified into Group A (red, n=18) or Group B (blue, n=92) based on the same criteria as in (A). Group A had significantly worse overall survival than Group B (5-year survival 73.2% vs. 86.2, respectively; P=0.0490, log-rank test).

Table 7. Analysis of Prognostic Factors for Disease-Specific Survival in CTEPH Survivors of Pulmonary Endarterectomy

Parameters	Univariate Cox proportional hazards analysis	P value
	HR　　95% CI
Data at diagnosis
Age	1.1675 (1.0544–1.3489)	0.0015
Sex (male)	– (　–　)
Coagulopathy	1.5982 (0.2951–8.6547)	NS
APA positive	2.4368 (0.2919–20.3396)	NS
WHO class III or IV	– (　–　)
D-dimer ≥1 μg/ml	0.3042 (0.0155–2.0648)	NS
Fibrinogen	1.0010 (0.9872–1.0133)	NS
TAT ≥3.0 ng/ml	– (　–　)
Factor VIII	1.0126 (0.9960–1.0329)	NS
TM	0.9849 (0.3317–1.6727)	NS
Plasminogen	0.9781 (0.9327–1.0293)	NS
PIC	0.6821 (0.0581–6.0548)	NS
PAI-1	0.9523 (0.6719–1.0999)	NS
α2PI	0.9689 (0.9182–1.0169)	NS
TPAIC	1.0044 (0.8218–1.1220)	NS
BNP	1.0014 (0.9974–1.0039)	NS
PaO2	0.9649 (0.9046–1.0310)	NS
Group A	8.8747 (1.4582–67.8643)	0.0195
mPAP	1.0856 (0.9974–1.1849)	NS
Cardiac index	0.7095 (0.1698–2.4270)	NS
PVR	1.2691 (0.9873–1.6750)	NS
Data at follow-up within 3 months
Postoperative mPAP	1.1171 (1.0371–1.2136)	0.0009
Postoperative cardiac index	0.3926 (0.0712–1.9317)	NS
Postoperative PVR	1.3529 (1.1284–1.6648)	0.0041
Over follow-up period
Use of pulmonary vasodilators	3.3622 (0.3899–29.3172)	NS

Continuous variables are indicated as per unit increase. HR of sex (male), WHO class (≥III), and TAT positivity (≥3.0 ng/ml) were not determined because of lack of patients with these characteristics who died of CTEPH. HR, hazard ratios. Other abbreviations as in Tables 1,2.

The 5-year overall survival rate of 110 patients including perioperative death and death from causes other than CTEPH at follow-up was 82.2%. Table S2B shows the characteristics of these patients stratified into Groups A (n=18) and B (n=92). Group A had significantly worse 5-year overall survival than Group B (73.2% vs. 86.2%, respectively; log-rank test P=0.0490; Figure B). The detailed causes of death of Group A were: disease-specific death (n=3, 16.7%), perioperative death (n=1, 5.6%) and other deaths (n=3, 16.7%), including cerebral hemorrhage (n=1), cerebral infarction (n=1) and multiple organ failure (n=1). For Group B, the causes of death were: disease-specific death (n=2, 2.2%), perioperative death (n=11, 12.0%) and other deaths (n=3, 3.3%), including interstitial pneumonitis (n=1), asphyxia (n=1), and natural causes (n=1) (data not shown).

Discussion

We found that the combination of high preoperative fibrinogen and low preoperative plasminogen values was a significant poor prognostic factor for postoperative CTEPH patients. In addition, we found that TM and plasminogen values were increased at follow-up after PEA; a high preoperative factor VIII value was associated with perioperative death; and D-dimer positivity at follow-up was a risk factor for residual PH. To our knowledge, this is the first study to show an association between the coagulation-fibrinolysis system and the long-term outcome of patients with CTEPH who undergo PEA, and to identify a risk factor for residual PH after PEA.

Several issues must be considered when interpreting these results. Group A, which included patients with high preoperative fibrinogen levels (≥291 mg/dl) together with low preoperative plasminogen activity (<100%), as well as high postoperative mPAP and high postoperative PVR, were associated with worse disease-specific survival after PEA for postoperative CTEPH. We previously reported that plasma fibrinogen and plasma plasminogen were independent prognostic factors in patients with medically treated CTEPH.¹¹ In addition, similar to this study, the combination of a fibrinogen level ≥ the median value of 291 mg/dl together with a plasminogen activity < the median of 101% was an extremely strong prognostic factor for survival in patients with medically treated CTEPH.¹¹ Although in this study, univariate analysis did not identify fibrinogen and plasminogen as significant predictors of disease-specific survival, high fibrinogen and low plasminogen levels tended to be associated with poor outcome. Taken together, the results of this and previous studies suggest that the long-term outcome of patients with CTEPH is likely to be affected by the coagulation-fibrinolysis system regardless of whether they are treated medically or surgically. Although the preoperative TPAIC levels in Group A were significantly higher than in Group B (Table S2A), univariate Cox proportional analysis did not identify TPAIC as a significant prognostic indicator of disease-specific survival (Table 7). Interestingly, the poor outcome of Group A was associated not only with disease-specific death but also with death from conditions other than CTEPH. The causes of nonspecific death in Group A included more cardiovascular disease and inflammatory disease. The chronic microvascular inflammation, as reflected by an increased plasma fibrinogen level,^15,16 might be associated with cerebrovascular events and multiple organ failure. In the present study, patients were classified into Group A or B, but patients could be further subdivided into 4 categories using median fibrinogen and plasminogen as cut-offs (subgroup 1, lower fibrinogen and higher plasminogen; subgroup 2, lower fibrinogen and lower plasminogen; subgroup 3, higher fibrinogen and higher plasminogen; subgroup 4, higher fibrinogen and lower plasminogen [=group A]). mPAP, CI and PVR were compared in these 4 subgroups by Kruskal-Wallis test. Among these 4 groups, there was no significant difference in baseline hemodynamics (data not shown). The survival analyses were performed in perioperative survivors for disease-specific survival (Figure S1A) and in all patients for overall survival (Figure S1B). In these 2 additional analyses, no significant differences were observed. However, there was a tendency for survival rate to be highest in subgroup 1, intermediate in subgroups 2 and 3, and lowest in subgroup 4 in both analyses.

Bonderman et al reported on prognostic factors other than hemodynamic parameters in patients with operative CTEPH.¹⁷ They concluded that PEA, PVR, and associated medical conditions such as splenectomy, ventriculoatrial shunt for the treatment of hydrocephalus, permanent central intravenous lines, inflammatory bowel disease, and osteomyelitis were predictors of survival. We had only 1 patient with an associated medical condition (ulcerative colitis), and the patient was classified into Group B. However, there have not been any studies that have exhaustively investigated the association between the outcomes of surgically treated patients with CTEPH and the parameters of the coagulation-fibrinolysis system. Our study provides evidence that coagulation-fibrinolysis parameters, especially fibrinogen and plasminogen values, may be useful for predicting the long-term outcomes of patients after PEA.

Coagulation is promoted by the conversion of fibrinogen into fibrin by thrombin.¹⁸ Therefore, an altered plasma fibrinogen level is considered to be an indicator of coagulopathy. Plasminogen is the inactive form of plasmin, which resolves fibrin blood clots. Therefore, altered plasminogen activity indicates aberrant fibrinolysis.¹⁹ Increased coagulation and decreased fibrinolysis may directly contribute to the progression of CTEPH, and also may result in death unrelated to CTEPH, such as death from a cerebrovascular event or renal or liver failure.

The beneficial effect of PEA on the fibrinolysis system was seen at follow-up in this study. The elevated TM levels after PEA were similar to the results of a previous study reported by Sakamaki et al.⁹ Additionally, this is the first study to report elevation of plasminogen after PEA. Our study, however, did not investigate the mechanisms of the increase in plasma TM and plasminogen values at follow-up after PEA. A hypothetical mechanism is that thrombi that result in consumed plasmin could be reduced by PEA, resulting in increased plasminogen. Considering that plasminogen has been reported to decrease with age,^11,20 the increase in plasminogen at follow-up is obviously affected not by variability over time but by PEA.

Univariate logistic regression analysis revealed that D-dimer positivity and elevated BNP in patients at the 1-year follow-up were associated with risk for residual PH. D-dimer and BNP assessments during the postoperative follow-up after PEA are probably useful noninvasive methods for predicting residual PH.

Elevated preoperative factor VIII activity was associated with perioperative death in our study. Elevated factor VIII activity was reported to be a risk factor for the occurrence of CTEPH²¹ and might reflect endothelial stress and damage. Severe endothelial damage may be associated with persistent PH after PEA, resulting in perioperative death.

Study Limitations

The study had several limitations. First, it was a retrospective study, and excluding potential confounding between the coagulation-fibrinolysis system and pulmonary hemodynamics is difficult. Second, deaths from conditions other than CTEPH were excluded from the disease-specific survival analysis and censored at that point; however, univariate Cox proportional analysis revealed that Group A (vs. B) remained significant when deaths other than CTEPH were considered as event counts (hazard ratio 7.2125 [95% confidence interval 2.1626–25.1331], P=0.0019; data not shown). Third, it was impossible to determine independent prognostic factors using multivariate analysis because of the improved long-term outcomes after PEA; only 6 of 102 patients died of CTEPH. Similarly, independent risk factors for residual PH at the 1-year follow-up were not identified because of the low number of study patients with residual PH. Fourth, some patients were receiving pulmonary vasodilators and all were receiving anticoagulant therapy at the time coagulation-fibrinolysis parameters were assessed. The medications might have affected these parameters. However, between the patients receiving or not receiving pulmonary vasodilators, there were no significant differences in the levels of coagulation-fibrinolysis markers, except for lower TM (4.7±2.6 vs. 6.1±3.0 FU/ml, P=0.0138) and lower TPAIC (11.9±9.7 vs. 14.0±7.0 ng/ml, P=0.0216) levels in patients receiving pulmonary vasodilators (data not shown). It is unclear if the reduced levels of fibrinolytic parameters in the patients receiving pulmonary vasodilators resulted from their clinical condition, which was severe enough to warrant the vasodilators, or from the pulmonary vasodilators. Regarding anticoagulant therapy, the proportion of patients within the therapeutic window for anticoagulant therapy (activated partial thromboplastin time ≥40 s, prothrombin time-international normalized ratio ≥1.5, or prothrombin time ≥18 s) was 78.3% (90 patients out of 115) at the time of preoperative blood sampling. However, the differences in the coagulation-fibrinolysis parameters of the patients within the therapeutic window and those not within the window were not significant (data not shown).

Conclusions

Preoperative high fibrinogen and low plasminogen values in patients with CTEPH who undergo PEA are associated with poor long-term postoperative outcome. PEA may benefit not only the pulmonary hemodynamics but also the fibrinolytic system of these patients.

Grants

This study was supported in part by grants from the Respiratory Failure Research Group of the Ministry of Health, Labor and Welfare of Japan, and the Pulmonary Hypertension Research Group of the Japan Agency for Medical Research and Development, AMED (K.T.), and a research grant from the Ministry of Education, Culture, Sports, Science and Technology of Japan (25461148) (N.T.).

Supplementary Files

Supplementary File 1

Table S1. Parameters of CTEPH patients assessed at diagnosis and at postoperative follow-up (<3 months after pulmonary endarterectomy) (n=48)

Table S2. (A) Characteristics and parameters of group A and B patients who survived surgery, (B) characteristics and parameters assessed at diagnosis of all group A and B patients

Figure S1. (A) Kaplan-Meier estimates of disease-specific survival in patients who survived pulmonary endarterectomy.

Please find supplementary file(s);

http://dx.doi.org/10.1253/circj.CJ-15-1208

References

• 1.    Jenkins  DP,  Madani  M,  Mayer  E,  Kerr  K,  Kim  N,  Klepetko  W, et al. Surgical treatment of chronic thromboembolic pulmonary hypertension. Eur Respir J 2013; 41: 735–742.
• 2.    Lang  IM,  Madani  M. Update on chronic thromboembolic pulmonary hypertension. Circulation 2014; 130: 508–518.
• 3.    Urushibara  T,  Tanabe  N,  Suda  R,  Kato  F,  Kasai  H,  Takeuchi  T, et al. Effects of surgical and medical treatment on quality of life for patients with chronic thromboembolic pulmonary hypertension. Circ J 2015; 79: 2696–2702.
• 4.    Jamieson  SW,  Kapelanski  DP,  Sakakibara  N,  Manecke  GR,  Thistlethwaite  PA,  Kerr  KM, et al. Pulmonary endarterectomy: Experience and lessons learned in 1,500 cases. Ann Thorac Surg 2003; 76: 1457–1464.
• 5.    Corsico  AG,  D’Armini  AM,  Cerveri  I,  Klersy  C,  Ansaldo  E,  Niniano  R, et al. Long-term outcome after pulmonary endarterectomy. Am J Respir Crit Care Med 2008; 178: 419–424.
• 6.    Ishida  K,  Masuda  M,  Tanabe  N,  Matsumiya  G,  Tatsumi  K,  Nakajima  N. Long-term outcome after pulmonary endarterectomy for chronic thromboembolic pulmonary hypertension. J Thorac Cardiovasc Surg 2012; 144: 321–326.
• 7.    Olman  MA,  Marsh  JJ,  Lang  IM,  Moser  KM,  Binder  BR,  Schleef  RR. Endogenous fibrinolytic system in chronic large-vessel thromboembolic pulmonary hypertension. Circulation 1992; 86: 1241–1248.
• 8.    Bonderman  D,  Turecek  PL,  Jakowitsch  J,  Weltermann  A,  Adlbrecht  C,  Schneider  B, et al. High prevalence of elevated clotting factor VIII in chronic thromboembolic pulmonary hypertension. Thromb Haemost 2003; 90: 372–376.
• 9.    Sakamaki  F,  Kyotani  S,  Nagaya  N,  Sato  N,  Oya  H,  Nakanishi  N. Increase in thrombomodulin concentrations after pulmonary thromboendarterectomy in chronic thromboembolic pulmonary hypertension. Chest 2003; 124: 1305–1311.
• 10.    Tanabe  N,  Kimura  A,  Amano  S,  Okada  O,  Kasahara  Y,  Tatsumi  K, et al. Association of clinical features with HLA in chronic pulmonary thromboembolism. Eur Respir J 2005; 25: 131–138.
• 11.    Kato  F,  Tanabe  N,  Urushibara  T,  Kasai  H,  Takeuchi  T,  Sekine  A, et al. Association of plasma fibrinogen and plasminogen with prognosis of inoperable chronic thromboembolic pulmonary hypertension. Circ J 2014; 78: 1754–1761.
• 12.    Moser  KM,  Auger  WR,  Fedullo  PF,  Jamieson  SW. Chronic thromboembolic pulmonary hypertension: Clinical picture and surgical treatment. Eur Respir J 1992; 5: 334–342.
• 13.    Riedel  M. Longterm follow-up of patients with pulmonary thromboembolism: Late prognosis and evolution of hemodynamic and respiratory data. Chest 1982; 81: 151.
• 14.    Ishida  K,  Masuda  M,  Tanaka  H,  Imamaki  M,  Katsumata  M,  Maruyama  T, et al. Mid-term results of surgery for chronic thromboembolic pulmonary hypertension. Interact Cardiovasc Thorac Surg 2009; 9: 626–629.
• 15.    Jennewein  C,  Tran  N,  Paulus  P,  Ellinghaus  P,  Eble  JA,  Zacharowski  K. Novel aspects of fibrin(ogen) fragments during inflammation. Mol Med 2011; 17: 568–573.
• 16.    Lowe  G,  Woodward  M,  Hillis  G,  Rumley  A,  Li  Q,  Harrap  S, et al. Circulating inflammatory markers and the risk of vascular complications and mortality in people with type 2 diabetes and cardiovascular disease or risk factors: The ADVANCE study. Diabetes 2014; 63: 1115–1123.
• 17.    Bonderman  D,  Skoro-Sajer  N,  Jakowitsch  J,  Adlbrecht  C,  Dunkler  D,  Taghavi  S, et al. Predictors of outcome in chronic thromboembolic pulmonary hypertension. Circulation 2007; 115: 2153–2158.
• 18.    Pizzo  SV,  Schwartz  ML,  Hill  RL,  McKee  PA. Mechanism of ancrod anticoagulation: A direct proteolytic effect on fibrin. J Clin Invest 1972; 51: 2841–2850.
• 19.    Castellino  FJ,  Ploplis  VA. Structure and function of the plasminogen/plasmin system. Thromb Haemost 2005; 93: 647–654.
• 20.    Kostka  T,  Para  J,  Kostka  B. Cardiovascular diseases (CVD) risk factors, physical activity (PA) and plasma plasminogen (Plg) in a random sample of community-dwelling elderly. Arch Gerontol Geriatr 2009; 48: 300–305.
• 21.    Kim  NH,  Lang  IM. Risk factors for chronic thromboembolic pulmonary hypertension. Eur Respir Rev 2012; 21: 27–31.