Persistent dyspnea complaints at long-term follow-up after an episode of
acute pulmonary embolism: Results of a questionnaire
F.A. Kloka,⁎, J.E. Tijmensena, M.L.A. Haecka, K.W. van Kralingenb, M.V. Huismana
aSection of Vascular Medicine, Department of General Internal Medicine-Endocrinology, Leiden University Medical Center, Leiden, The Netherlands
bDepartment of Pulmonary Medicine, Leiden University Medical Center, Leiden, The Netherlands
Received 3 September 2007; received in revised form 27 December 2007; accepted 17 February 2008
Available online 22 April 2008
Background: There is a lack of information on long term complications of patients with pulmonary embolism (PE), including chronic
complaints of dyspnea.
Methods: Consecutive patients with a prior diagnosis of acute PE and an age and gender matched control group with no medical history of PE
were presented with a questionnaire, designed to establish the presence, severity and possible causes of dyspnea in the clinical course of PE.
Results: The questionnaire was taken in 48 PE-survivors 40±7.4 months after PE; 27 patients (56%) had complaints of dyspnea. Sixteen
(35%) were categorized as NYHA class II, 6 (13%) as class III and 5 (10%) as class IV. Overall, 19 patients (70%) had new or worsened
complaints after PE. The study included 61 controls. Corrected for gender, age and medical history, the control group was significantly less
dyspnoeic compared to the PE survivors (pb0.001). Corrected for gender and age, patients were 4 times more often in NYHA class II (OR
3.6 95%CI 1.4–9.7) and 7-fold more often in NYHA class III or IV (OR 6.5 95%CI 1.7–24), both compared to control subjects.
Conclusion: A large percentage of patients with prior PE have persistent complaints of dyspnea at long term follow-up. The majority of them
developed new or worsened dyspnea after the thrombo-embolic event. In comparison to a control population without a medical history of VTE,
© 2008 European Federation of Internal Medicine. Published by Elsevier B.V. All rights reserved.
Keywords: Pulmonary embolism; Follow-up; Questionnaire; Dyspnia
There is extensive literature on pulmonary embolism (PE)
focusing on its incidence, prevention, diagnostic strategies
and treatment. Long term prospective studies on the clinical
course of PE are however lacking. Patients with a first epi-
, dependent on the underlying cause of thromboembolism.
After this, patients are usually no longer subject to clinical
supervision. In spite of frequent clinical impression of per-
sisting dyspnea, no study has systematically evaluated this
phenomenon. Most follow-up studies have dealt with differ-
ent diagnostic strategies, recurrent thrombotic events, major
bleeding from anticoagulant therapy and mortality [2–6].
Little studies have follow-up for more than 2 years and des-
cribe the natural course of dyspnea complaints [7–9]. Rele-
vance of persisting dyspnea complaints is underscored by
Pengo et al , describing an incidence of 4% chronic
thromboembolic pulmonary hypertension (CTEPH) in pa-
tients after a first episode of PE. Chronic thromboembolic
pulmonary hypertension is a late complication of PE .
an average interval of 17.6 months . A total of 65 patients
European Journal of Internal Medicine 19 (2008) 625–629
Abbreviations: PE, Pulmonary embolism; PH, Pulmonary hypertension;
CTEPH, Chronic thrombo-embolic pulmonary hypertension; NYHA, New
York Heart Association; CT, Computed tomography; OR, Odds Ratio;
APPT, Activated partial thromboplastin time; ANOVA, Analysis of
variance; CI, Confidential interval.
⁎Corresponding author. LUMC (C4-70), Albinusdreef 2, Postbus 9600,
2300 RC Leiden, The Netherlands. Tel.: +3171 5262085; fax: +3171
E-mail address: F.A.Klok@LUMC.nl (F.A. Klok).
0953-6205/$ - see front matter © 2008 European Federation of Internal Medicine. Published by Elsevier B.V. All rights reserved.
died immediately or within hours after diagnosis, 6 died
during follow-up, 1 was lost due to missing contact speci-
fications and 1 refused informed consent. The total follow up
concerned 68 patients; of those 10 (15%) had severe dyspnea
and a further 9 (13%) remained moderately dyspnoeic.
In two of the patients with severe dyspnea, pulmonary
hypertension (PH) with right-ventricular failure was clearly
Standardized clinical follow-up, e.g. in the form of a vali-
dated questionnaire, could help the physician identifying
patients with chronic complaints due to PE. In these latter
patients, the cause of chronic symptoms should be further
evaluated. Unfortunately, there are no reported studies in
which patients with PE were followed up to screen them for
long term complaints of dyspnea.
We designed a questionnaire to establish the presence,
severity and possible causes of dyspnea in the clinical course
of PE and presented it to a group of patients with a history of
PE and a control group of random selected persons without a
history of PE.
2. Materials and methods
This newly developed questionnaire in Dutch, consisting
patients in our clinic of vascular medicine. It was designed to
assess complaints of dyspnea by establishing the presence,
severity, impact on daily activities and possible causes of
dyspnea in patients with a history of PE. Patients were asked
for medical history of cardiac or pulmonary disease, anemia,
use of medication and smoking behaviour. Ten questions
were asked to distinguish the severity of the complaints.
Exercise was classified into none (rest), mild (walking short
distances and personal hygiene), moderate (normal daily
activities, housekeeping) and heavy exercise (swimming,
running). Also, a few questions were meant to identify signs
of PHand right heart failure (e.g. syncope and angina).When
dyspnea was confirmed, three questions were asked to deter-
mine the course of the dyspnea; whether the complaints ex-
isted before, whether the complaints had worsened and
the thrombo-embolic event. A history of cardiopulmonary
disease was defined as any current or former visit to a cardio-
logist or pulmonary physician or current or former use of
Finally, the questions were used to classify patients ac-
cording to the criteria of the New York Heart Association
(NYHA) functional classification, although not strictly appli-
has been designed for qualifying exertional intolerance in
stages of heartfailure and has been used to classify patients
with PH . Also, NYHA classification was used as ins-
trument of evaluation inlong term follow-up after pulmonary
2.2. Patients and taking the questionnaire
Eighty-eight consecutive patients, as part of a large
diagnostic multicenter study , with a diagnosis of PE by
CT-scan, diagnosed in our hospital between November 2002
and August 2004 were selected for the present study. Our
study population comprised of patients with either a first or
subsequent episode of venous thromboembolism after a
follow-up of at least 2 years. They were treated according to
1.5 to 2.5 prolongation of the APTT (activated partial throm-
boplastin time), followed by oral anticoagulant therapy for
6 months. Medical records were searched for contact speci-
fications and documented clinical course.
The questionnaire was taken by telephone in all surviving
patients. Possible answers were ‘yes’ and ‘no’. The response
with a medical history of pulmonary embolism to persons
with no such medical history, a control cohort was created by
taking the questionnaire in randomly selected persons in a
shopping street in the same geographical region during 2
consecutive days. They were asked to answer all questions
and their medical history, as previously defined, was noted.
Only persons with a history of pulmonary embolism were
The scoring of the NYHA class was performed by two
independent researchers without knowledge of patient or
control status. In case of disagreement, a third researcher was
consulted. All patients and controls consented to participate
in the study and the Medical Ethical Committee of our hos-
pital approved our protocol.
2.3. Statistical analysis
Confidence intervals were calculated by the modified
Wald method, the Fisher’s exact test was used to compare
discrete variables. Fisher’s analysis of variance (ANOVA)
was used to test differences among the study- and control
population. Data were considered significant at pb0.05.
SPSS (SPSS for windows 12.0.1, Inc. 1989–2003) and
GraphPad (GraphPad software, Inc. 2002–2005) software
were used for statistical calculations.
3.1. Patients’ characteristics
Forty-eight patients completed the questionnaire; they
comprised the study population. Their age at time of PE was
questionnaire was 40±7.4 (24–51) months. Twenty-three
patients (32%) had died before the questionnaire could be
taken (Fig. 1). In seventeen of the remaining patients the
questionnaire could not be taken due to refused cooperation
(3), inaccessibility (11) or absent contact specifications (3).
The flowchart of patient selection is displayed in Fig. 1. To
626F.A. Klok et al. / European Journal of Internal Medicine 19 (2008) 625–629
create a control cohort, 90 random selected subjects were
presented with the questionnaire. The response rate in the
control group was 68%. In most cases, lack of time to com-
plete the questionnaire was indicated as the main reason for
refusal. The study included 61 controls; their age was 54±15
(25–88). The control group did not differ from the PE-group
in gender, age, history of cardio-pulmonary history and smo-
king habits (Table 1).
3.2. Outcome questionnaire
dyspnea (Table 2). Sixteen of 48 (35%) were categorized as
class IV (Table 2). The periodof time between the PE and the
questionnaire had no correlation with the severity of com-
plaints. In 19 of the 27 patients reporting dyspnea (70%), the
complaints were experienced as new or worsened after PE.
The control group comprised of 61 persons, of which 47
(77%) were categorized as NYHA I, 10 (16%) as NYHA II
and 4 (6.6%) as NYHA III (Table 2). No controls were clas-
sified as NYHA IV. Corrected for gender and age, the control
group was significantly less dyspnoeic than the PE survivors
(pb0.001). After stratification for medical history, in both
groups with (p=0.012) or without (pb0.001) cardio-pulmo-
nary comorbidity, controls were significantly less dyspnoeic.
Corrected for gender and age, patients were classified 4-fold
more often in NYHA class II (OR 3.6 95%CI 1.4–9.7) and 7-
24), both compared with control subjects.
Twelveof16(75%) patients withaprior historyofcardiac
or pulmonary disease reported dyspnea. Thirty-two patients
did not have a cardiac or pulmonary history; in this latter
group 15 (47%) had complaints of dyspnea. Significantly
more patients with cardio-pulmonary history were dyspnoeic
(p=0.02) and they were classified in a higher NYHA-class
(pre-) syncope, angina or palpitations as new or worsened
symptoms after PE were mentioned. When compared to the
cohort without newly developed symptoms of right heart
failure, these 9 patients were classified in a higher NYHA-
We studied the natural course of dyspnea complaints in
PE-patients by estimating the incidence and severity of per-
sisting dyspnea symptoms using a standardized question-
naire. Our data demonstrates a surprisingly high number of
patients with complaints of dyspnea after a mean period of
40 months after PE. The severity of their complaints varied
between mild limitations in physical activity to inability to
thirds of the patients with dyspnea indicated their complaints
to have started or had worsened after the PE.
in addition to a considerable contribution of (partly not iden-
tified) cardio-pulmonary comorbid conditions, incomplete
resolution of pulmonary emboli contributes to the deve-
lopment of dyspnea. A recently published systematic review
study  suggests persistent perfusion defects in more than
50% of PE patients 6 months after diagnosis, after which
resolution of thrombi appears to reach a plateau phase. Clots
can be classified in to several categories: central fillingdefect
or complete occlusion, eccentric clot contiguous with the
vessel wall, filling defect with central contrast material indi-
catingrecanalization andsevere arterial luminal narrowingor
vessel occlusion of a stenosed artery . Persistent thrombi
cause partial or complete obliteration of pulmonary arteries
which may result in increased ventilation–perfusion mis-
match and dead space ventilation. Indeed, dead space venti-
lation as determined by V/Q-scans has been shown to corre-
late with lung perfusion defects in subjects with acute
PE . Also, the development of ventilation-perfusion
Characteristics of 48 studied patients; N=number, μ=mean, σ=standard
General characteristics of 48 studied patients and 61 controls
Sex [F/M] N (%)
Age at PE [years] μ±σ (min–max)
Time between PE and questionnaire
[month] μ±σ (min–max)
Age at questionaire μ±σ (min–max)
Medical history of cardio-pulmonary
disease N (%)
Smoking N (%)
N=number, μ=mean, σ=standard deviation.
Fig. 1. Flowchart of patient selection.
627 F.A. Klok et al. / European Journal of Internal Medicine 19 (2008) 625–629
inequality has been demonstrated to provide a major contri-
bution to the deranged gas exchange seen following pulmo-
nary embolization . In addition, survivors of PE present
signs of ventilation to perfusion mismatch at exertion
3.1 years on average after the thromboembolic event .
No study has yet been performed to combine complaints of
dyspnea with dead space ventilation measurements and pul-
monary perfusion tests after long term follow-up in patients
with acute PE.
One other notorious complication of unresolved thrombi
after PE is CTEPH . Characteristically, the main sym-
ptom ofCTEPH isexertional dyspnea andsigns ofrightheart
failure . The natural history of CTEPH is not completely
known as a result of late presentation in the course of disease
. Originally, it was assumed that 0.1%–0.5% of patients
surviving PE develop CTEPH . Recent published studies
find incidences of 1.0% after one year  and 1.0, 3.1 en
3.8% after 6 months, 1 and 2 years  subsequent to PE.
Notably, both Becattini et al  and Pengo et al  did not
find any new cases of CTEPH after 2 years. This challenges
to years before symptoms appear . In our study nine pa-
tients had reported clinical signs of right heart failure. These
patients had significantly more severe dyspnea (p=0.01) and
more cardiopulmonary comorbidity (p=0.049). Importantly,
history takingof these patients revealedpreviousechocardio-
graphic work up in only four of them.
Our study has strengths and limitations. Even with a res-
ponse rate of 74%, we studied a relatively limited number of
patients over a period of time, partly due to a mortality rate of
32%. As result, the confidence limits of our observations are
wide. Second, the PE-cohort was questioned by phone
whereas the controls were questioned in person. This could
questions with only a possible “yes” or “no” answer. Because
the results of the questionnaire were interpreted by two inde-
pendent researchers without knowledge of patient or control
status, we do not consider this to have influenced are results.
Third, about 33%of thepatients hada prior history ofcardiac
or pulmonary disease. Obviously, this has great influence on
cardiopulmonary disease, significantly more patients had
dyspnea and complaints were overall more serious compared
to patients with no history of cardio-pulmonary disease. Al-
though we found a comparable prevalence of prior or con-
current cardiovascular disease in both study groups, it can be
reasoned that sampling on the street causes a bias towards
more healthy individuals. Finally, because our study-end-
point was the outcome of the questionnaire, we were not
capable of determining the causes of the dyspnea.
In conclusion, we found 56% of patients with prior PE to
event. The majority of them developed new or worsened
dyspnea. In comparison to a control population without a
medical history of VTE, PE patients were overall signifi-
cantly more dyspnoeic. Larger, prospective clinical outcome
studies using cardiopulmonary exercise testing and echocar-
diography are however needed to determine the cause of the
dyspnea in this patient population.
 Büller HR, Agnelli C, Hul RD, et al. Antithrombotic therapy for
venous thromboembolic disease the seventh ACCP conference on
antithrombotic and thrombolytic therapy. Chest 2004;126:401S–28S.
 Nijkeuter M, Sohne M, Tick LW, et al. The natural course of hemo-
dynamically stable pulmonary embolism. Clinical Outcome and Risk
Factors in a Large Prospective Cohort Study, vol. 131. Chest; 2007.
 Perrier A, Roy PM, Sanchez O, et al. Multidetector-row computed
tomography in suspected pulmonary embolism. N Engl J Med 2005;
 Nendaz MR, Bandelier P, Aujesky D, et al. Validation of a risk
score identifying patients with acute pulmonary embolism, who are
at low risk of clinical adverse outcome. Thromb Haemost 2004;91:
 Le Gal G, Righini M, Roy PM, et al. Prediction of pulmonary
embolism in the emergency department:the revisedGeneva score. Ann
Intern Med 2006;144:165–71.
 van Belle A, Büller HR, Huisman MV, et al. Effectiveness of managing
suspected pulmonary embolism using an algorithm combining clinical
probability, D-dimer testing, and computed tomography. JAMA 2006;
 Pengo V, Lensing AW, Prins AW, et al. Incidence of chronic thromo-
embolic pulmonary hypertension after pulmonary embolism. N Engl J
 Riedel M, Stanek V, Widimsky J, et al. Longterm follow-up of patients
with pulmonary thromboembolism. Late prognosis and evolution of
hemodynamic and respiratory data, vol. 81. Chest; 1982. p. 151–8.
 Phear D. Pulmonary embolism. A study of late prognosis, vol. 2.
Lancet; 1960. p. 832–5.
 Hoeper MM, Mayer E, Simonneau G, et al. Chronic thromboembolic
pulmonary hypertension. Circulation 2006;113:2011–20.
Outcome questionnaire: NYHA classification in study population and control cohort; effect of PE on their complaints
NYHA classification of 48 patients with a history of pulmonary embolism and gender and age matched control group (N=61) with no such medical history
n (%) 95% CIn (%) 95% CIn (%) 95% CIn (%) 95% CI
Overall PE population
Complaints worsened after PE
Complaints same after PE
628 F.A. Klok et al. / European Journal of Internal Medicine 19 (2008) 625–629
 Rich S. Primary pulmonary hypertension: executive summary from
the World Symposium on Primary Pulmonary Hypertension. Evian,
France: WHO; 1998.
 Barst RJ, McGoon M, Torbicki A, et al. Diagnosis and differential
assessment of pulmonary arterial hypertension. J Am Coll Cardiol
 Habicht JM, Hämerli R, Perruchoud A, et al. Long-term follow-up in
pulmonary embolectomy: is NYHA (dyspnea) classification reliable?
Eur J Cardio-thorac Surg 1996;10:32–7.
 Doerge HC, Schoendube FA, Loeser H, et al. Pulmonary embolect-
omy: review of a 15-year experience and role in the age of throm-
bolytic therapy. Eur J Cardiothorac Surg 1996;10:952–7.
 Bauer EP, Laske A, von Segesser LK, et al. Early and late results after
surgery for massive pulmonary embolism. Thorac Cardiovasc Surg
 Meyer G, Tamisier D, Sors H, et al. Pulmonary embolectomy: a
20-year experience at one center. Ann Thorac Surg 1991;51:
 Nijkeuter M, Hovens MM, Davidson BL, et al. Resolution of throm-
boemboli in patients with acute pulmonary embolism: a systematic
review. Chest 2006;129:192–7.
 Kline JA, Kubin AK, Patel MM, et al. Alveolar dead space as a pre-
dictor of severity of pulmonary embolism. Acad Emerg Med 2000;7:
 Dantzker DR, Bower JS. Alterations in gas exchange following pul-
monary thromboembolism. Chest 1982;81:495–501.
 Ciurzyński M, Kurzyna M, Bochowicz A, et al. Long-term effects of
acute pulmonary embolism on echocardiographic Doppler indices and
functional capacity. Clin Cardiol 2004;27:693–7.
 Fedullo PF, Auger WR, Kerr KM, et al. Chronic thromboembolic
pulmonary hypertension. N Engl J Med 2001;345:1465–72.
 Moser KM, Auger WR, Fedullo PF. Chronic major-vessel thromboem-
bolic pulmonary hypertension. Circulation 1990;81:1735–43.
 Becattini C, Agnelli G, Pesavento R, et al. Incidence of chronic throm-
boembolic pulmonary hypertension after a first episode of pulmonary
embolism. Chest 2006;130:172–5.
629 F.A. Klok et al. / European Journal of Internal Medicine 19 (2008) 625–629