The distance between the vein and lesions predicts the requirement of carina ablation in circumferential pulmonary vein isolation

Abstract

Additional ablation in the pulmonary vein (PV) carina region is sometimes required to achieve electrical isolation following circumferential pulmonary vein isolation (PVI). This study investigated the procedural predictors for the requirement of additional carina ablation to achieve complete electrical isolation with PVI.
Eighty patients with drug-refractory paroxysmal AF underwent circumferential PVI. After the first round of PVI, we placed circular catheters inside the veins to identify the residual PV potentials, and also performed electroanatomic mapping to observe the earliest activation sites during sinus rhythm. The requirement of an additional gap and carina ablation, and the optimal distance that predicted an incomplete PV block were assessed. In the first 40 patients, 43% of the ipsilateral PVs were electrically isolated after the initial PVI. Subsequent ablation of the gaps and ablation of the carina were required in the remaining 57% PVs. The only predictor of the requirement of carina ablation was the mean distance between the lesion-related scar and the ostia (P = 0.03). The longer the distance from the isolating lesions to the PV ostia (>8 mm) predicted an incomplete PV isolation after the first round of circumferential isolation. In the next 40 patients, a fixed distance of 8 mm to the PV ostia decreased the requirement of a carina ablation and resulted in a shorter procedure time (P < 0.05).
This study indicated the importance of complete linear lesions and additional carina ablation when the wide area circumferential PV isolation was applied.

Full text

Risk stratification and clinical outcomes in patients with acute
pulmonary embolism☆
Chi-Ming Huang
a
, Yen-Chung Lin
c
, Yenn-Jiang Lin
a,b,
⁎, Shih-Lin Chang
a,b
, Li-Wei Lo
a,b
, Yu-Feng Hu
a,b
,
Chern-En Chiang
a,b
, Kang-Ling Wang
a,b
, Shih-Ann Chen
a,b
a
Division of Cardiology, Department of Medicine, Taipei Veterans General Hospital, Taipei, Taiwan
b
Institute of Clinical Medicine, and Cardiovascular Research Center, National Yang-Ming University, Taipei, Taiwan
c
Department of Internal Medicine, Taipei Medical University Hospital, Taipei, Taiwan
abstractarticle info
Article history:
Received 15 February 2011
Received in revised form 13 June 2011
Accepted 15 June 2011
Available online 24 June 2011
Keywords:
Mortality
Prognosis
Pulmonary embolism
Risk stratification
Hemodynamic instability
Objectives: Pulmonary embolism is a common disease associated with a high mortality rate. The risk
assessment and appropriate treatment selection of patients with acute pulmonary embolism remains a
challenge.
Design and methods: This single center cohort study included a total of 150 patients (96 male, age =71 ±
15 years) with acute pulmonary embolism confirmed by spiral-computed tomography or magnetic resonance
image. The prognostic performance of the clinical characteristics and laboratory values were investigated to
predict the in-hospital hemodynamically instable events and 30-day all-cause mortality.
Results: The rate of in-hospital hemodynamic instability and 30-day all-cause mortality was 21% and 12%,
respectively. A multivariate Cox regression analysis demonstrated that a heart rate ≥110 bpm (odd ratio 4.26
[95% CI 1.42–12.77]), chronic pulmonary disease (6.47 [1.99–21.04]), WBC ≥11,000 mm
3
(3.78 [1.32–10.82]),
and D-dimer level≥4.0 μg/mL (3.68 [1.01–13.43]) independently predicted the 30-day fatal outcome. A
Kaplan–Meier survival analysis showed that the categorization based on the number of risk factors was
significantly associated with the likelihood of 30-day all-cause mortality (P b0.0001).
Conclusions: The initial presentation of tachycardia, presence of chronic pulmonary disease, elevated WBC
and D-dimer on admission can be used to identify the risk for a short-term fatal outcome within 30 days in
patients with acute pulmonary embolism.
© 2011 The Canadian Society of Clinical Chemists. Published by Elsevier Inc. All rights reserved.
Introduction
Pulmonary embolism which encompasses a wide spectrum of
illnesses, along with the complex progress and diverse prognosis,
remains one of the leading causes of morbidity and mortality in the
emergency and cardiovascular setting [1]. The risk of death is
particularly high during the acute phase and then decreases over time.
Previous studies demonstrated that the overall in-hospital mortality in
patients with acute pulmonary embolism was 3% within 48 h and 9.6%
during hospitalization [2,3]. Approximately 5–20% of patients develop
circulatory shock or respiratory failure [4]. Therefore, the early
identification of the patients with pulmonary embolism at risk of
developing hemodynamic instability or death is a major issue. It is
important to stratify the patients at high risk for death who should
receive a specific therapeutic management. Several clinical features and
diagnostic tests observed at the time of the diagnosis could predict a
worse outcome from pulmonary embolism [5]. However, those studies
varied widely in the predictive value of the clinical outcome, and
required many comprehensive studies during the hospitalization. This
study was intended to provide a fast risk stratification method at the
initial presentation. The objective of this study wasto test the prognostic
power of the baseline laboratory values and clinical parameters at the
initial presentation for predicting the short-term hemodynamic
instability and 30-day all-cause mortality.
Methods
Study population
This study enrolled 150 consecutive patients hospitalized in Taipei
Veterans General Hospital, a tertiary transferal medical center in
Taipei City, between December 1, 2004 and September 31, 2009, All
patients were admitted to the emergency department for acute
pulmonary embolism, which was diagnosed by computed tomogra-
phy angiography (CTA). Patients aged 18 years or older with
objectively confirmed acute pulmonary embolism were considered
eligible. Both hemodynamically stable and instable patients on
Clinical Biochemistry 44 (2011) 1110–1115
☆No conflicts of interest of all authors.
⁎Corresponding author at: Division of Cardiology, Taipei Veterans General Hospital,
201, Sec. 2, Shih-Pai Road, Taipei, Taiwan. Fax: + 886 2 2873 5656.
E-mail address: linyennjiang@gmail.com (Y.-J. Lin).
0009-9120/$ –see front matter © 2011 The Canadian Society of Clinical Chemists. Published by Elsevier Inc. All rights reserved.
doi:10.1016/j.clinbiochem.2011.06.077
Contents lists available at ScienceDirect
Clinical Biochemistry
journal homepage: www.elsevier.com/locate/clinbiochem

admission who could be stabilized after the initial medical treatment
were enrolled into this study. The following patients were excluded:
1) recent acute coronary syndrome in the previous 6 months; 2)
significant septic condition; 3) illness with a predicted 6-month
mortalityN50% (e.g., terminal metastatic cancer condition, end-stage
acquired immune deficiency syndrome, end-stage heart or renal
failure with no plan for transplantation or hemodialysis therapy); 4) a
do-not-resuscitate order with a clinical plan to not treat the patient
for pulmonary embolism.
Data acquisition
This retrospective study was approved by the Ethics Committee by
hospital Ethics Committee. Complete information on the clinical course,
diagnostic and therapeutic management, and nursingfollow-up records
were reviewed. Data were collected on (1) the clinical symptoms and
signs of the patients at diagnosis; (2) the presence of underlying
diseases or predisposing factors for pulmonary embolism (hyperten-
sion, diabetes, hypercholesterolemia, heart failure, coronary artery
disease, cancer, chronic pulmonary disease, immobilization, and recent
major surgery requiring general anesthesia); (3) the findings of all
diagnostic procedures and laboratory data performed on admission,
including a complete blood cell count, blood glucose, C-reactive protein,
arterial blood gas, D-dimer (Sysmex CA-7000, Siemens Germany), and
cardiac enzymes levels, chest roentgenograms, 12-lead eletrocardio-
graphy, echocardiography, computed tomographic studies, and ultra-
sound tests for deep vein thrombosis; (4) the treatment given to the
patients (heparin, low-molecular-weight heparin, warfarin, thrombol-
ysis, surgical pulmonary embolectomy, catheter thrombus fragmenta-
tion, and caval filter implantation); and (5) the in-hospital clinical
course of the patients, including any in-hospital hemodynamic
instability or bleeding complications. The survival was confirmed
through the use of the medical and nursing records.
The shock index (a ratio of the heart rate to systolic blood pressure). A
shock index ≥1 has been shown for predicting patients with a high risk of
an adverse outcome. This ratio has been shown to be related to the in-
hospital mortality and it is sensitive for predicting a poor prognosis alone
or in combination with trans-thoracic echocardiography [6].
Definition of the clinical end points
The overall mortality within 30 days was defined as the primary
endpoint. Hemodynamic deterioration during hospitalization was
taken into account to evaluate the in-hospital clinical course of the
patients with acute pulmonary embolism, including: (1) new onset of
hemodynamic collapse, (2) need for treatment upgrading, such as
thrombolysis therapy, emergency surgical embolectomy or catheter
thrombus fragmentation, (3) need for endotracheal intubation or
cardiopulmonary resuscitation, (4) systolic blood pressure persis-
tently less than 100 mm Hg, refractory to volume loading, and
requiring vasopressors treatment.
Patient follow-up
All patients were treated with warfarin during the follow-up
periods. All patients were under regular clinical follow-up at 0.5, 1, 2,
3 months, and every 3 months after discharge. At the end of the
follow-up period, all included patients were contacted by telephone
and interviewed by one of the study coordinators who were blinded
to the results of biochemical analyses. Mortality was assessed using
patient or proxy interviews and/or hospital chart review. Two
independent experts adjudicated the cause of death as definite fatal
PE or death from other causes. In necessary, a staff assistant contacted
the survivors by telephone approximately 3 months later.
Statistical analysis
Statistical analysis was performed using the SPSS statistical
package (version 17.0, Chicago, IL). Quantitative variables are
expressed as the mean ±standard deviation. A Chi-square test with
a Fisher's exact test was used for the categorical data. The normally
distributed continuous variables were compared using the Student's
t-test, whereas the abnormally distributed variables were compared
using the Mann–Whitney Utest. Variables selected to be tested in the
multivariate analysis were those with a P value of b0.2 in the
univariate models. Logistic regression and Cox-regression analysis
was applied for the multivariate analysis for the in-hospital morbidity
Table 1
Baseline and clinical characteristics.
Variables (N=150)
Men, (%) 64%
Age, (years) 71.3±14.8
Body mass index, mean (kg/m
2
) 24±4.8
Cigarette smoking, (%) 25.3%
Initial vital signs
Systolic blood pressure (mm Hg) 128±27.8
Heart rate, mean (bpm) 99.6±21.8
Body temperature, (mean ±SD), °C 36.6±0.96
Past medical history
Hypertension, (%) 47.3%
Diabetes mellitus, (%) 20.0%
Hypercholesterolemia, (%) 13.3%
Congestive heart failure, (%) 15.3%
Coronary artery disease, (%) 16.7%
Cancer, (%) 30.7%
Chronic pulmonary disease, (%) 14.0%
History of venous thromboembolism, (%) 10.0%
Recent major surgery, (%) 6.7%
Initial presentation symptoms
Dyspnea, (%) 80.7%
Chest pain, (%) 34.0%
Syncope, (%) 13.3%
Hemodynamic instability during hospitalization, (%) 21.3%
Value = mean ±SD.
Table 2
The baseline non-invasive diagnostic findings.
Electrocardiography Rhythms
Sinus rhythm (%) 80.7%
Atrial fibrillation (%) 19.3%
Right ventricular strain patterns
S1Q3T3,% 18.7%
CRBBB/ICRBBB (%) 13.3%
Inverted T-waves in V1 through V3, (%) 22.7%
Laboratory data White blood cell count, cumm 10,227 ±5759
Hemoglobin (g/dL) 12.7±2.2
Platelet (10,000 mm
3
)22±11
Glucose (mg/dL) 148.2±93.2
C-reactive protein (mg/dL) 5.4±7.0
CK (IU/L) 74.4±125.2
CKMB (IU/L) 5.2±5.9
Troponin I (ng/mL) 0.63± 2.13
D-dimer (μg/mL) 10.0±21.3
PaO
2
(mmHg) 98.6±64.9
Echocardiography LA diameter, mm 39.2 ±8.3
RV dilatation or failure,% 42.7%
LV ejection fraction,% 50.7±12.1
Deep vein thrombosis shown
by ultrasound,%
32.0%
Abbreviations: CRBBB = complete right bundle branch block; CK = Creatine kinase;
CKMB = myocardial-specific isoenzyme of creatine kinase (MB form); PaO
2
= partial
pressure of oxygen in arterial blood; LA = left atrium; RV = right ventricle; LV = left
ventricle.
1111C.-M. Huang et al. / Clinical Biochemistry 44 (2011) 1110–1115

and 30-day mortality respectively. A Kaplan–Meier analysis was used
to investigate the multivariate survival predictive power of a model
including four independent factors with relation to the mortality
within 30 days.
Results
Clinical presentation
The baseline characteristics and results of the noninvasive
diagnostic tests are summarized in Tables 1 and 2. Among the
patients included, the mean age was 71.3 ±14.8 years, and 64% were
men (Table 1). A history of cancer was noted in 30.7% of the patients,
hypertension in 47.3%, coronary artery disease in 16.7%, heart failure
in 15.3%, and chronic pulmonary disease in 14%. A total of 10% of the
patients had previous history of venous thromboembolism. Only 10
patients (6.7%) had submitted to recent major surgeries. The most
common symptom at presentation was dyspnea (80.7%).
Electrocardiographic findings of sinus rhythm were found in 80.7%,
and atrial fibrillation in 19.3%. Findings of right ventricular strain
patterns, such as an S1-Q3-T3 pattern, inverted T-waves in leads V1–
V3, and right-bundle branch block pattern were observed in 18.7%,
22.7% and 13.3%, respectively. All patients had documented pulmo-
nary artery embolism by CT scanning (149 patients, 99%) or magnetic
resonance imaging of the chest (1 patient, 1%). The incidence of
concomitant deep vein thrombosis proved by ultrasound was 32%.
In-hospital course and predictors of in-hospital hemodynamic instability
All patients (100%) were treated in the acute phase with
conventional therapeutic doses of unfractionated heparin or low-
molecular-weight heparin. The mean duration of hospitalization of
overall 150 patients with acute pulmonary embolism was
20.4 ± 14.8 days. Seven patients (4.7%) received thrombolytic therapy
during hospitalization. A surgical embolectomy was performed in 2
patients (1.3%). During the hospitalization, 32 patients (21.3%)
developed hemodynamic instability, including 26 patients (17.3%)
who developed respiratory failure requiring mechanical ventilation
and 23 patients (15.2%) developing shock requiring vasopressor
treatment. Overall mortality within 30 days occurred in 18 patients
Table 3
Univariate and multivariate predictors of in-hospital hemodynamic instability.
Variables Univariate analysis P Multivariate analysis P
OR (95% CI) OR (95% CI)
Age 1.009 (0.982–1.038) 0.512 –
Male 1.500 (0.238–9.441) 0.666 –
Body mass index 0.982 (0.893–1.080) 0.711 –
Cigarette smoking 1.323 (0.515–3.397) 0.561 –
Initial vital signs
Heart rate 1.023 (1.004–1.042) 0.019 1.009 (0.980–1.039) 0.542
Systolic blood pressure 0.980 (0.965–0.996) 0.014 0.997 (0.969–1.025) 0.825
Shock indexN1 (HR/SBP) 4.331 (1.735–10.810) 0.002 2.237 (0.372–13.451) 0.379
Body temperature 0.793 (0.514–1.222) 0.293 –
Eletrocardiogaphy findings
Sinus tachycardia 2.265 (0.988–5.189) 0.053 –
S1Q3T3 1.459 (0.504–4.222) 0.486 –
CRBBB/ICRBBB 1.543 (0.538–4.427) 0.420 –
Inverted T-waves in V1 through V3 1.515 (0.563–4.079) 0.411 –
Atrial fibrillation 2.155 (0.874–5.313) 0.095 –
Past medical history
Hypertension 1.201 (0.547–2.637) 0.647 –
Diabetes 1.455 (0.577–3.667) 0.427 –
Hypercholesterolemia 1.714 (0.601–4.891) 0.314 –
Congestive heart failure 1.785 (0.663–4.804) 0.251 –
Coronary artery disease 1.556 (0.586–4.132) 0.375 –
Cancer 1.425 (0.586–3.465) 0.435 –
Chronic pulmonary disease 2.080 (0.760–5.693) 0.154 –
History of venous thromboembolism 1.390 (0.411–4.696) 0.596 –
Recent major surgery 1.640 (0.399–6.739) 0.492 –
Immobilization 10.733 (2.594–44.404) 0.001 9.840 (1.754–55.201) 0.009
Initial presentation symptoms
Dyspnea 9.644 (1.259–73.875) 0.029 2.523 (0.280–22.716) 0.409
Chest pain 1.440 (0.644–3.219) 0.374 –
Syncope 4.909 (1.826–13.201) 0.002 5.000 (1.351–18.503) 0.016
Initial laboratory values
WBC, cumm 1.000 (1.000–1.000) 0.074 1.000 (1.000–1.000) 0.318
Hemoglobin, g/dL 1.040 (0.872–1.240) 0.661 –
Glucose, mg/dL 1.004 (1.000–1.008) 0.056 1.004 (0.999–1.009) 0.165
Total cholesterol, mg/dL 0.984 (0.974–0.996) 0.006 –
C reactive protein, mg/dL 1.024 (0.972–1.078) 0.372 –
D-dimer, μg/mL 1.037 (1.007–1.068) 0.015
D-dimer≧4.0 μg/mL 4.800 (1.819–12.664) 0.002 4.199 (1.253–14.074) 0.020
CK, IU/L 1.003 (0.999–1.006) 0.138 –
Troponin I, ng/ml 1.317 (0.862–2.012) 0.202 –
PaO
2
, mm Hg 1.002 (0.997–1.008) 0.420 –
DVT on sonography 1.045 (0.451–2.424) 0.918 –
Abbreviations: DVT= deep vein thrombosis; HR = heart rate; SBP = systolic blood pressure.
1112 C.-M. Huang et al. / Clinical Biochemistry 44 (2011) 1110–1115

(12%) with a hospitalization of 22.5±20.3 days. Among all of the 18
patients died within 30 days, 12 patients (66.7%) died of acute
pulmonary embolism, 3 (16.7%) patients died of underlying cancer
and 3 (16.7%) patients died of severe septic condition.
In the univariate analysis (Table 3), the patients who developed
hemodynamic instability during hospitalization tended to have a
lower systolic blood pressure on admission, higher heart rate on
admission, history of recent immobilization more than 4 days and
symptoms of syncope while presenting. In the ECG findings, the
rhythms and depolarization changes did not predict the outcome. In
the initial laboratory data, a lower cholesterol level, high D-dimer
level, and troponin-I level correlated with the hemodynamic
instability. The other clinical findings, laboratory data, and prevalence
of concomitant deep vein thrombosis were similar between the two
groups.
In the multivariate logistic regression analysis (Table 3), initial
presentation of symptoms of syncope (OR: 5.0, p=0.016), a previous
history of immobilization (OR: 9.84, p =0.009), and a D-dimer
level≥4.0 μg/mL on admission (OR: 4.199, p= 0.02) independently
predicted the hemodynamic instability during hospitalization. A
shock indexN1 on admission was insignificant (PN0.05).
Risk stratification of 30-day overall mortality
As shown in Table 4, patients with a fatal outcome within 1 month
had a higher heart rate, history of chronic pulmonary disease, and
laboratory data with higher WBC and D-dimer levels on admission.
The development of in-hospital hemodynamic instability events was
the most significant predictor in the univariate analysis (HR =11.792,
Pb0.0001). The other clinical and laboratory parameters did not
correlate with the 30-day all-cause mortality (P =NS).
In the multivariate Cox regression model, 4 predictors are shown
to be correlated with the 30-day mortality independently, including a
heart rate≥110 bpm, chronic pulmonary disease, white blood
cell≥11,000 mm
3
and D-dimer level ≥4.0 μg/mL (Table 5). The
patients with acute pulmonary embolism were further categorized
into three risk groups according to the number of the previous proved
independent predictors. A Kaplan–Meier survival analysis revealed
the prediction of the all-cause mortality by the number of prognostic
markers (Fig. 1). The 30 day all-cause mortality was 2.4%, 14.9% and
69.2%, in the low (≤1 risk factor), intermediate (2 risk factors) and
high risk (≥3 risk factors) groups, respectively.
Table 4
Risk factors for the 30-day all-cause mortality (Univariate Cox Regression analysis).
Variable HR (95% CI) P-value
Male 1.234 (0.128–11.870) 0.856
Age 0.999 (0.969–1.031) 0.963
Body mass index 0.974 (0.868–1.092) 0.647
Cigarette smoking 0.364 (0.083–1.604) 0.182
Initial vital signs
Heart rate 1.031 (1.010–1.053) 0.004
HR≧110 bpm 3.641 (1.411–9.396) 0.008
SBP 1.000 (0.983–1.018) 0.967
Shock index N1 (HR/SBP) 2.504 (0.934–6.712) 0.068
Body temperature 0.982 (0.601–1.605) 0.942
ECG findings
Sinus tachycardia 1.414 (0.548–3.647) 0.474
S1Q3T3 0.464 (0.107–2.019) 0.306
CRBBB/ICRBBB 1.105 (0.320–3.818) 0.874
Inverted T-waves in V1 through V3 1.137 (0.405–3.189) 0.807
Atrial fibrillation 1.559 (0.556–4.375) 0.399
Past medical history
Hypertension 1.074 (0.675–1.709) 0.764
Diabetes 1.435 (0.688–2.992) 0.336
Hypercholesterolemia 0.720 (0.413–1.256) 0.247
Congestive heart failure 1.163 (0.337–4.016) 0.812
Recent major surgery 1.136 (0.414–3.115) 0.804
History of venous thromboembolism 1.413 (0.515–3.872) 0.502
Cancer 2.360 (0.937–5.946) 0.069
Chronic pulmonary disease 3.303 (1.239–8.805) 0.017
Immobilization 0.750 (0.360–1.564) 0.443
Initial laboratory values
Glucose 1.002 (0.998–1.005) 0.409
CRP 1.03 (0.983–1.079) 0.220
WBC ≧11,000 cumm 3.157 (1.223–8.144) 0.017
Hemoglobin 0.852 (0.687–1.056) 0.145
D-dimer 1.013 (1.005–1.022) 0.003
D-dimer≧4.0 μg/mL 4.479 (1.287–15.590) 0.018
CK 1.001 (0.999–1.004) 0.339
CKMB 1.027 (0.961–1.098) 0.427
Troponin I 1.017 (0.826–1.252) 0.873
PaO
2
1.002 (0.996–1.008) 0.442
Initial presentation symptoms
Dyspnea 4.365 (0.581–32.802) 0.152
Chest pain 1.992 (0.791–5.018) 0.144
Syncope 1.928 (0.634–5.857) 0.247
Hemodynamic instability
during hospitalization
11.792 (4.196–33.139) b0.0001
Deep vein thrombosis in ultrasound 1.029 (0.386–2.742) 0.954
Fig. 1. The four risk factors with significance from the multivariate Cox regression
analysis of the 30-day all-cause mortality were used for a Kaplan–Meier survival curve
estimation, including an HR ≥110 bpm, WBC≥11,000 mm
3
, D-dimer level ≥4.0 μg/mL
and chronic pulmonary disease. The patients with pulmonary emboli are categorized
into three groups. The blue line indicates the 30-day survival curve of the patients with
none or one risk factor. The green line represents the survival curve of the patients with
two risk factors. The yellow line was the survival curve of the patients with three or four
risk factors. The differences between each group are all statistically significant.
Table 5
Multivariate Cox Regression analysis of the 30-day all-cause mortality.
Variate HR (95%CI) P-value
HR≧110 bpm 4.257 (1.419–12.774) 0.010
Chronic pulmonary disease 6.470 (1.989–21.044) 0.002
WBC≧11,000 mm
3
3.779 (1.320–10.815) 0.013
D-dimer≧4.0 μg/mL 3.681 (1.009–13.428) 0.048
Shock indexN1 1.719 (0.601–4.914) 0.312
Cancer 2.442 (0.872–6.837) 0.089
1113C.-M. Huang et al. / Clinical Biochemistry 44 (2011) 1110–1115

Discussion
Main findings
This study examined the utility of the initial clinical presentation,
past co-morbidities, and laboratory tests results on admission for
predicting the development of in-hospital hemodynamic instability
events and the 30-day mortality. First, an initial presentation of
syncope, history of immobilization and D-dimer level ≥4.0 μg/mL, had
prognostic utility for the prediction of in-hospital hemodynamic
instability events. Second, an HR≥110 bpm on admission, past
history of chronic pulmonary disease, WBC count ≥11,000 mm
3
on
admission, and D-dimer level≥4.0 μg/mL on admission could predict
the 30-day survival. The patients with acute pulmonary embolism
were further categorized into a risk stratification system according to
how many of the 4 independent predictors of the 30-day all-cause
mortality they had. This quick scoring method stratified the patients
with acute pulmonary embolism into three risk groups and could
significantly predict the 30-day survival.
Prognostic stratification based on the vital signs
Previous studies showed that the hemodynamic presentation was
associated with final fatal outcome. In the International Cooperative
Pulmonary Embolism Registry (ICOPER) Study, the mortality rate was
15.1% and 58.3% in patients with and without a stable hemodynamic
presentation, respectively [7]. Similar results were observed in this
study, and the 30-day mortality rate in the patients who did not
develop any hemodynamically instable events during the hospitali-
zation was only 4.2%, and was up to 40.6% in the patients with in-
hospital hemodynamic instability events that expired within 30 days
(pb0.001). The Pulmonary Embolism Prognostic Index (PESI) is the
clinical scoring system for identifying patients with an increased risk
of adverse outcomes [8]. Among the 10 predictive factors in the PESI,
the HR and an HR≥110 bpm could independently predict the 30-day
all-cause mortality. The parameter of a shock index N1 had an
insignificant predictive value for the survival within 30 days
(p=0.068). However, both the systolic blood pressure and shock
index did not predict the fatal outcome within 30 days in this study.
There are two possible explanations for this finding when compared
to the previous studies. First, this study did not include hemodynam-
ically unstable patients with immediate mortality or surgical
intervention. The other possible reason is our study patients were
from a single tertiary-referring medical center in Asia. Difference in
the ethnical factors and associated morbidities existed between our
study population and those of the other previous studies.
Prognostic stratification based on the clinical history and presenting
symptoms
The predictive value of the clinical history for the short-term 30-
day fatal outcome in the patients with acute pulmonary embolism
remained inconsistent in the previous studies [3,7,9,10]. An age over
70 years, history of bed rest for 5 days or more, chronic pulmonary
disease, cancer and renal failure were shown to be risk factors for
mortality in the patients with pulmonary embolism in the ICOPER
registry [7]. A North American large observational study (Heit Cohort
study) also reported similar independent predictors of the short-term
outcome, based on the history and clinical symptoms [9]. On the other
hand, the MAPPET registry demonstrated that the role of COPD or
heart failure in the prediction of the mortality was not confirmed in a
multivariate analysis [3]. In a study presented by Geibel et al., no
difference in the past history between the survivors and non-
survivors was observed with regard to the history of a recent surgery
or major trauma, previous venous thromboembolic events, stroke,
cancer or pregnancy [10,11].
Regarding the symptoms, syncope was a predictor of in-hospital
hemodynamic instability but not related to the 30 days all cause
mortality in this study. This is compatible with a retrospective review
of 154 consecutive patients with acute pulmonary embolism, in which
the patients with and without syncope had similar epidemiological
and clinical features (including respiratory failure, right heart failure
and arterial hypotension), and hospital mortality [11]. In this study,
only a history of chronic pulmonary disease had a strong predictive
value of the 30-day all-cause mortality and without the occurrence of
in-hospital hemodynamic instability in our study. Recent study
demonstrated that pulmonary embolism may be one of the causes
of exacerbation of COPD [12]. On the other hand, the COPD could be a
high risk for PE due to a variety of factors including limited mobility,
inflammation, and comorbidities. In this study, 66.7% of the patients
with COPD with mortality was due to respiratory failure. It remained
to be seen whether early aggressive intervention may reduce the risk
of mortality of these patients.
Prognostic stratification based on the change in the ECG
ECG signs suggestive of right ventricular strain (S1Q3T3 pattern,
right-bundle branch block, pulmonary P waves, T-waves inversion in
the right precordial leads) have been shown to correlate with the
extent of the perfusion defect as assessed by lung scans or pulmonary
angiography [13]. The impact of the individual ECG findings on
admission on the 30-day mortality was evaluated in 508 patients with
acute pulmonary embolism [10]. Atrial arrhythmias, complete right-
bundle branch block, peripheral low voltage, pseudo-infarction
pattern in leads III and aVF, and ST segment changes (elevation or
depression) over the left precordial leads, were all significantly more
frequent in the patients who had a fatal outcome. A number of ECG
findings have been shown to be associated with adverse outcome
events in the short-term course of patients with pulmonary
embolism, in particular T-wave inversion in precordial leads [14,15].
In this study, none of the ECG signs were related to the short-term
hemodynamic instability or 30-day mortality. There were explana-
tions for this finding. The prognostic value of the ECG has usually been
demonstrated for abnormalities of a recent onset of disease. Further,
the ECG signs suggestive of right ventricular strain required a prior
ECG for comparison, and that may not be available in the initial phase
of the patient management and could limit the prognostic
implications.
Prognostic stratification based on the laboratory biomarkers
The D-dimer levels appeared to be associated with the extent of
the thromboembolic burden in patients with pulmonary embolism in
several previous studies. Ghanima et al. showed that the D-dimer
value was related to the severity of the pulmonary embolism assessed
by various radiological, biochemical and clinical markers and might
have had a potential value as a prognostic marker for the severity of
pulmonary embolism [16]. Patients with pulmonary embolism who
had D-dimer levels below 1.5 μg/mL had a very low mortality in a
study published in 2006 [17]. In that study, significantly higher rates
of in-hospital hemodynamic instability events and a 30-day all-cause
mortality were found in the patients with a D-dimer level ≥4.0 μg/mL
which had been confirmed as an independent predictor of an adverse
outcome in the multivariate Cox-regression analysis. The level of the
D-dimer seems to be a useful predictor of the future outcomes for
patients with acute pulmonary embolism. High troponin-I was
another potent predictor of 30-day all cause mortality [22], however,
in systemic review and meta-analysis [23], troponin-I failed to discern
the risk of mortality for patients with acute pulmonary embolism. In
our study, high troponin-I levels were not independent predictor of
patient mortality, but correlated with hemodynamics instability.
1114 C.-M. Huang et al. / Clinical Biochemistry 44 (2011) 1110–1115

Regarding the white blood cell count, a previous study demon-
strated that leukocytosis may be associated with pulmonary embo-
lism [18]. In a subgroup analysis of the RIETE registry, cancer patients
with acute venous thromboembolism and elevated WBC count had an
increased incidence of venous thromboemboli recurrences, major
bleeding or death. A multivariate analysis confirmed that an elevated
WBC count was independently associated with an increased incidence
of all three complications [19]. There have been no previous studies
that have demonstrated the prognostic value of an elevated WBC
count for the future outcome. The WBC count (≥11,000) was proven
to be independently predictive of the 30-day all-cause mortality in our
multivariate Cox-regression analysis.
Clinical implications
Traditionally, risk stratification in patients with acute pulmonary
embolism is made on a clinical basis. In recent years, right ventricular
dysfunction assessed on transthoracic echocardiography had been
described as one of the strongest predictors of an early mortality with
nonmassive pulmonary embolism [20]. However, many smaller hospi-
tals do not have accredited echocardiographic laboratories, and most
large academic hospitals do not have a uniform availability of
echocardiography. A survey in 2003 found that only 5% of emergency
physicians at large, academic, teaching hospitals with cardiology
fellowships reported the ability to obtain echocardiography within 2 h
at all times [21]. This study was intended to provide a fast risk
stratificationmethod for patients with acute pulmonary embolism in the
acute phase, especially when echocardiography cannot be available. Our
findings were obtained in patients with stable and instable hemody-
namic conditions while accounting for a very wide range of covariates.
In this study, the patients with an acute pulmonary embolism were
categorized into three risk groups according to the four independent
predictors of the 30-day mortality, including an HR ≥110 bpm,
chronic pulmonary disease, WBC count≥11,000 mm
3
and D-dimer
level≥4.0 μg/mL. This study demonstrated a rapid risk stratification
model using the easily available clinical and laboratory parameters on
admission for patients with acute pulmonary embolism.
Limitations
First, the mortality rate of our study population was higher than
some previous studies. This hospital was the tertiary referral medical
center, and most of the patients were referred from other hospitals.
The incidence of malignancy is higher in our patients. 33% (50)
patients were aged more than 80 years. The result reflects the real
condition in our hospital. Second, the time frame considered for
performing blood samples, as well as the range of times used to select
the highest value of cardiac biomarkers, were chosen arbitrarily.
Further studies are needed to address this issue. Third, pro B-Type
Natriuretic Peptide was not measured in all patients and this may
have affected the results of this marker.
Conclusion
Risk stratification is the cornerstone of the modern management of
acute pulmonary embolism. The four clinical and laboratory factors
predicting the 30-day all-cause mortality identified in this study
(history of chronic pulmonary disease, and a heart rate≥110 bpm,
WBC count≥11,000 mm
3
and D-dimer level≥4.0 μg/mL on admis-
sion) could be routinely identified and could therefore easily be
included in a risk stratification scheme in daily practice.
References
[1] Torbicki A, Galié N, Covezzoli A, Rossi E, De Rosa M, Goldhaber SZ, et al. Right heart
thrombi in pulmonary embolism: results from the International Cooperative
Pulmonary Embolism Registry. J Am Coll Cardiol 2003;41:2245–51.
[2] Kline JA, Hernandez-Nino J, Newgard CD, Cowles DN, Jackson RE, Courtney DM,
et al. Use of pulse oximetry to predict in-hospital complications in normotensive
patients with pulmonary embolism. Am J Med 2003;115:203–8.
[3] Konstantinides S, Geibel A, Olschewski M, Heinrich F, Grosser K, Rauber K, et al.
Association between thrombolytic treatment and the prognosis of hemodynam-
ically stable patients with major pulmonary embolism: results of a multicenter
registry. Circulation 1997;96:882–8.
[4] Kucher N, Printzen G, Goldhaber SZ. Prognostic role of brain natriuretic peptide in
acute pulmonary embolism. Circulation 2003;107:2545–7.
[5] Wood KE. Major pulmonary embolism: review of a pathophysiologic approach to
the golden hour of hemodynamically significant pulmonary embolism. Chest
2002;121:877–905.
[6] Otero R, Trujillo-Santos J, Cayuela A, Rodríguez C, Barron M, Martín JJ, et al.
Haemodynamically unstable pulmonary embolism in the RIETE Registry: systolic
blood pressure or shock index? Eur Respir J 2007;30:1111–6.
[7] Goldhaber SZ, Visani L, De Rosa M. Acute pulmonary embolism: clinical outcomes
in the International Cooperative Pulmonary Embolism Registry (ICOPER). Lancet
1999;353:1386–9.
[8] Aujesky D, Obrosky DS, Stone RA, Auble TE, Perrier A, Cornuz J, et al. A prediction
rule to identify low-risk patients with pulmonary embolism. Arch Intern Med
2006;166:169–75.
[9] Heit JA, Silverstein MD, Mohr DN, Petterson TM, O'Fallon WM, Melton III LJ, et al.
Predictors of survival after deep vein thrombosis and pulmonary embolism: a
population-based, cohort study. Arch Intern Med 1999;159:445–53.
[10] Geibel A, Zehender M, Kasper W, Olschewski M, Klima C, Konstantinides SV, et al.
Prognostic value of the ECG on admission in patients with acute major pulmonary
embolism. Eur Respir J 2005;25:843–8.
[11] Calvo-Romero JM, Pérez-Miranda M, Bureo-Dacal P. Syncope in acute pulmonary
embolism. Eur J Emerg Med 2004;11:208–9.
[12] Rizkallah J, Man SF, Sin DD. Prevalence of pulmonary embolism in acute
exacerbations of COPD: a systematic review and metaanalysis. Chest 2009;135:
786–93.
[13] Stein PD, Dalen JE, McIntyre KM, Sasahara AA, Wenger NK, Willis III PW, et al. The
electrocardiogram in acute pulmonary embolism. Prog Cardiovasc Dis 1975;17:
247–57.
[14] Jimenez D. ECG for risk stratification in patients with pulmonary embolism. Eur
Respir J 2005;26:366–7 author reply.
[15] Toosi MS, Merlino JD, Leeper KV. Electrocardiographic score and short-term
outcomes of acute pulmonary embolism. Am J Cardiol 2007;100:1172–6.
[16] Ghanima W, Abdelnoor M, Holmen LO, Nielssen BE, Ross S, Sandset PM, et al. D-
dimer level is associated with the extent of pulmonary embolism. Thromb Res
2007;120:281–8.
[17] Aujesky D, Roy PM, Guy M, Cornuz J, Sanchez O, Perrier A, et al. Prognostic value of
D-dimer in patients with pulmonary embolism. Thromb Haemost 2006;96:
478–82.
[18] Afzal A, Noor HA, Gill SA, Brawner C, Stein PD, et al. Leukocytosis in acute
pulmonary embolism. Chest 1999;115(5):1329–32.
[19] Trujillo-Santos J, Di Micco P, Iannuzzo M, Lecumberri R, Guijarro R, Madridano O,
et al. Elevated white blood cell count and outcome in cancer patients with venous
thromboembolism. Findings from the RIETE Registry. Thromb Haemost 2008;100:
905–11.
[20] Goldhaber SZ. Echocardiography in the management of pulmonary embolism.
Ann Intern Med 2002;136:691–700.
[21] Jones AE, Kline JA. Availability of technology to evaluate for pulmonary embolism
in academic emergency departments in the United States. J Thromb Haemost
2003;1:2240–2.
[22] Jiménez D, Díaz G, Molina J, Martí D, Del Rey J, García-Rull S, et al. Troponin I and
risk stratification of patients with acute nonmassive pulmonary embolism. Eur
Respir J 2008;31(4):847–53.
[23] Jiménez D, Uresandi F, Otero R, Lobo JL, Monreal M, Martí D, et al. Troponin-based
risk stratification of patients with acute nonmassive pulmonary embolism:
systematic review and metaanalysis. Chest 2009;136(4):974–82.
1115C.-M. Huang et al. / Clinical Biochemistry 44 (2011) 1110–1115