Article Text

Download PDFPDF
Original article
Functional and haemodynamic assessment of mild-to-moderate pulmonary valve stenosis at rest and during exercise
  1. Pieter De Meester1,
  2. Roselien Buys2,
  3. Alexander Van De Bruaene1,
  4. Charlien Gabriels1,
  5. Jens-Uwe Voigt1,
  6. Luc Vanhees2,
  7. Paul Herijgers3,
  8. Els Troost1,
  9. Werner Budts1
  1. 1Department of Cardiology, University Hospitals Leuven, Leuven, Belgium
  2. 2Department of Rehabilitation Sciences, University of Leuven, Leuven, Belgium
  3. 3Department of Cardiac Surgery, University Hospitals Leuven, Leuven, Belgium
  1. Correspondence to Werner Budts, Department of Cardiology, University Hospitals Leuven, Leuven, Belgium, Herestraat 49, Leuven B-3000, Belgium; werner.budts{at}


Objective In adult patients with mild-to-moderate pulmonary valve (PV) stenosis, exercise capacity and haemodynamics have not been extensively studied, although regular exercise is recommended. Therefore, we aimed to assess exercise capacity to study the increase in PV gradient during exercise and to evaluate the impact of this increased pressure load on the RV.

Methods Nineteen patients (8 female; 29±6.4 years) with isolated mild-to-moderate PV stenosis and no prior cardiac interventions were consecutively enrolled from the outpatient clinic of adult congenital heart disease. All patients underwent cardiopulmonary exercise testing, transthoracic echocardiography and bicycle stress echocardiography. Results for exercise testing were compared with age-matched and gender-matched control patients.

Results In the studied population, resting heart rate (89±11 vs 75±14 bpm; p=0.001), peak power (199±66 vs 263±68 W; p=0.006); peak VO2 (31.2±9.9 vs 39±7.4 mL/kg/min; p=0.011); oxygen uptake efficiency slope (2430±913 vs 3292±943(mL/min)/(L/min); p=0.007) and VE/VCO2 slope (26.8±5.2 vs 22.6±4.3; p=0.01) differed significantly from controls. A linear increase of peak PV gradient with increasing flow was observed in the pooled dataset (Pearson's R=0.947; p<0.0001) and slopes identical as for control patients were obtained for the oxygen pulse–workload relationship. Right heart morphology and function were preserved in the studied patients.

Conclusions Patients with mild-to-moderate PV stenosis have decreased exercise capacity. A linear increase in PV gradient with flow suggests a fixed valve area throughout the exercise. Although systolic RV pressure load increases during exercise, good ventricular performance was observed without signs of functional or morphological changes of the right heart.

Clinical trial number: NCT01444222

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.


Pulmonary valve (PV) stenosis occurs in 1 per 2000 live births and accounts for 8% of all congenital heart defects. In 80–90% of cases, the valve is dome-shaped with a pinhole orifice and no or incomplete separation of the leaflets. Rarely, the malformation consists of thickened valve leaflets with commissural fusion or PV dysplasia.1 ,2 ,3 Evaluation of severity of stenosis is made by Doppler echocardiography with a transvalvular gradient <36 mm Hg, between 36–64 and >64 mm Hg considered as mild, moderate and severe PV stenosis, respectively.4

In cases of severe outflow tract obstruction, infundibular hypertrophy and small subendocardial infarctions in the right ventricular free wall and papillary muscles can occur. Consequently, exercise capacity is reduced due to an inability to increase stroke volume with the increase in heart rate being the sole contributor to an increase in cardiac output.5 ,6 Conversely, the natural evolution of mild PV stenosis is considered to be benign without progression of the valve gradient after adolescence and no deaths are observed during an 8-year follow-up period.3 ,7 Because of this, no restrictions in physical activity are imposed.8 ,9

However, the exercise capacity and haemodynamics of patients with mild-to-moderate PV stenosis have not been investigated thoroughly. Although the defect is simple, the increase in valvular gradient at peak exercise and its impact on exercise performance has hardly been investigated. Romeih et al10 made a first attempt to investigate the mechanisms underlying possible impairment of the exercise capacity in patients with moderate PS. They demonstrated in a small group of asymptomatic patients with native moderate PS that the exercise capacity and cardiac reserve remained normal. However, their evaluation of cardiac function was based on pharmacological stress, rather than physical stress, and the impact on PV and right ventricular function was not documented.

Therefore, we wanted to assess exercise capacity in patients with mild-to-moderate PV stenosis to study the increase in PV gradient with increasing cardiac output and to evaluate the impact of the pressure load at peak exercise on the RV.


Patient selection and data collection

From 18 October 2011 until 24 May 2013, consecutive patients were included from the outpatient clinic of congenital heart disease at the University Hospital of Leuven. Patients with isolated mild-to-moderate PV stenosis, as evidenced by a PV Doppler gradient <64 mm Hg and no prior cardiac interventions, were eligible for inclusion in the study. Exclusion criteria were age under 16 years and contraindication for exercise testing. Furthermore, patients with known coronary artery disease, significant valvular disease other than pulmonary regurgitation (>1/4), associated cardiac malformations and a history of arrhythmias or current arrhythmias were excluded from the study. Patients with Noonan's syndrome or other chromosomal abnormalities were excluded. All patients had the classical form of PV stenosis. All patients underwent symptom-limited cardiopulmonary exercise testing and bicycle stress echocardiography. In all patients, informed consent was obtained at inclusion. Approval for patient inclusion and data collection was obtained from the local ethics committee.

Cardiopulmonary exercise testing

All participants underwent a symptom-limited, incremental cardiopulmonary exercise test on a bicycle ergometer (Ergometrics, 800S, Ergometrics, Bitz, Germany). Workload was increased by 20 W every minute until exhaustion.

Simultaneous gas exchange measurement, 12-lead ECG monitoring (Marquette Max Personal Exercise Testing, Wisconsin, USA) and blood pressure monitoring were performed. The oximeter was calibrated before each test. Respiratory data were collected by breath-by-breath analysis (Oxygen Alpha, Jaeger, Mijnhardt, Bunnik, The Netherlands) with continuous monitoring of oxygen and carbon dioxide concentrations in the inspired and expired air to determine oxygen uptake (VO2) and carbon dioxide output (VCO2). Peak oxygen consumption (peakVO2) was defined as the highest 30 s average of oxygen uptake at the end of the test. Respiratory gas exchange ratio was calculated as carbon dioxide output divided by oxygen uptake (VCO2/VO2). Anaerobic threshold was determined by the V-slope method. All slopes were calculated after exclusion of the first minute of exercise. The VE/VCO2 slope was calculated by plotting the VCO2 as a function of ventilation. The measurements after the respiratory compensation point were omitted. The oxygen uptake efficiency slope (OUES) was calculated by plotting the VO2 as a function of the log of ventilation. VO2 was plotted as a function of work rate resulting into the VO2/WR relationship. Oxygen pulse was calculated as the VO2 divided by heart rate.11

Control population

To compare the gas-exchange parameters of the patient population, a control population was selected from the database of cardiopulmonary exercise testing at the university hospitals of Leuven by 1:1 age and gender matching. Matching was automated by using the ‘Fuzzy’ extension command as available from the SPSS website ( Parameters were set to obtain exact matches.

Transthoracic echocardiography and bicycle stress echocardiography

All patients underwent a standard transthoracic echocardiographic examination. All examinations were performed on a Vivid 9 ultrasound system (General Electric Vingmed Ultrasound, Horten, Norway) equipped with a 3 MHz probe. Exercise echocardiography was performed on a semi-supine ergometer tilted laterally from 20° to 30° to the left (Easystress, Ecogito, Medical sprl, Liège, Belgium). The protocol started at 25 W with an increment of 25 W every 2 min until the maximum tolerated load.

A single observer performed the measurements. All measurements were made in triplicate and averaged for analysis. Analysis was done offline using dedicated software (EchoPac, Genereral Electric Vingmed Ultrasound, Horten, Norway).

In all patients, a complete resting echocardiographic study was performed in the supine position. All Doppler echocardiographic and tissue Doppler imaging record­ings were obtained during normal respiration. At rest, valvular insufficiency of the atrioventricular valves was evaluated semiquantitatively by colour Doppler flow mapping and valvular stenosis was evaluated by aligning the continuous wave Doppler beam with the studied valve. The highest velocities obtained were included for analysis, and the pressure gradient was calculated from the simplified Bernouillie equation. From the apical window, the pulsed Doppler sample volume was placed at the tips of the mitral valve to evaluate the mitral valve inflow pattern. Also, right ventricular area was measured at end-diastole and end-systole from a RV focused apical four-chamber view. Right atrium area was measured from the four-chamber view at end-systole. Right ventricular function was evaluated by measuring the tricuspid annular plane systolic excursion (TAPSE) on M-mode echocardiography and fractional area change (FAC=(end-diastolic area – end-systolic area)/end-diastolic area) at rest from a RV-focused apical four-chamber view.

At rest as well as at every stage during exercise, cardiac output and PV gradient were obtained. Cardiac output was calculated from the flow velocity time integral in the left ventricular outflow tract obtained by pulsed-wave Doppler echocardiography. The left ventricular outflow tract diameter was measured at rest and assumed constant throughout exercise. PV velocities were measured by aligning the CW Doppler beam along the PV and the pressure gradient was subsequently calculated.

Statistical analysis

Data of continuous variables are presented as means±SD and categorical variables as frequencies and percentages.

Demographic and exercise parameters of the study and control population were evaluated using the independent Student t test and χ2 test where appropriate.

Pearson's correlation coefficient was calculated for resting PV gradient and PV gradient at peak exercise in each subject. Pearson's correlation coefficient for cardiac output and PV gradient was calculated on the pooled measurements of all patients and every stage of exercise. Furthermore, the slope of the O2pulse–workload relationship of patients was compared with that of controls. To allow for analysis of the pooled data and account for the intersubject variability, Poon recalibration of the data was performed.12

All tests were two-tailed. A p value <0.05 was considered significant. Analyses were performed using SPSS (V.22 SPSS, Chicago, USA).


Patient characteristics

Nineteen patients with PV stenosis, 8 (42%) female, aged 29±6.4 years, were included in the study. None of the patients had known chromosomal abnormalities, and all were diagnosed with the classic form of PV stenosis. Demographic data can be found in table 1.

Table 1

Demographics of patients with pulmonary valve stenosis and healthy controls

Cardiopulmonary exercise test

All patients performed a cardiopulmonary exercise test at near maximal levels as evidenced by respiratory exchange ratio of 1.2±0.08 in the patient population and 1.2±0.07 in the control population (p=0.067).

In the studied patient population, heart rate increased from 89 bpm±11 to 181 bpm±11. Resting heart rate was significantly higher in the patient population compared with age-matched and gender-matched controls (89±11 vs 75±17 bpm; p=0.001). Heart rate reserve was significantly lower in the patient population ((92±16 vs 110±21 bpm; p=0.007) (figure 1). However, peak HR was not different compared with controls (181±11 vs 187±17 bpm; p=0.5).

Figure 1

Comparison of exercise parameters for patients with pulmonary valve stenosis and the age-matched and gender-matched control population. (A) Evaluation of heart rate at rest, peak exercise and the resultant heart rate reserve (HRR). (B) Peak oxygen uptake (VO2), peak power and power at anaerobic threshold (AT). and (C) Ventilatory efficiency by oxygen uptake efficiency slope (OUES) and ventilation/carbon dioxide output (VE/VCO2) slope.

Compared with controls, patients with PV stenosis presented with significantly lower exercise capacity as evidenced by a lower peak power (199±66 vs 263±68 W; p=0.006); peak VO2 (31.2±9.9 vs 39±7.4 mL/kg/min; p=0.011) and power at anaerobic threshold (83±34 vs 146±41 W; p<0.0001) (figure 1).

Furthermore, patients had evidence of lower ventilatory efficiency as shown by a significantly lower OUES (2430±913 vs 3292±943 (mL/min)/(L/min); p=0.007) and higher VE/VCO slope (26.8±5.2 vs 22.6±4.3; p=0.01) (figure 1). No significant difference in peak oxygen pulse (14.66±4.3 vs 15.98±4.31; p=0.351) or in VO2/WR relationship (9.47±1.18 vs 9.73±1.13; p=0.491) was observed.

Lastly, the slope of the oxygen pulse as function of increasing workload did not differ between the patient population and the control population (y=0.039x+6.72 and 0.040x+5.61, respectively; p=0.525) (figure 2).

Figure 2

Pooled analysis of all obtained measurements of oxygen pulse as a function of increasing workload. No difference in the slope of the oxygen pulse–work load relationship was found.

Stress echocardiography

Mean power at peak exercise for semi-supine bicycle stress echocardiography was 146 W±47. The echocardiographic parameters at rest and during exercise of the PV stenosis patients can be found in table 2. LVEF was 68%±9%. None of the patients had signs of left ventricular diastolic dysfunction. Peak gradient across the PV was 24 mm Hg±12 at rest and increased to 54 mm Hg±22 at peak exercise. Cardiac output increased from 5.1 L/min±1.09 to 12.4 L/min±3.2. Maximal gradient across the PV correlated strongly with resting pulmonary gradient (Pearson's R=0.905; p<0.0001) (figure 3). Evaluation of linearity of the pooled data showed a strong correlation between cardiac output and PV gradient (y=4.162x − 2.463; Pearson's R=0.947; p<0.0001) (figure 4).

Table 2

Echocardiographic evaluation at rest and at peak exercise in the patient group

Figure 3

Analysis of pulmonary valve gradient at rest and at peak exercise for all subjects. PV, pulmonary valve.

Figure 4

Pooled analysis of all obtained measurement of cardiac output and pulmonary valve gradient during exercise echocardiography.

PV gradient at rest and at peak exercise respectively did not correlate with peak cardiac output (R=−0.068; p=0.781 and R=0.212; p=0.480), peak stroke volume (R=−0.177; p=0.469 and R=0.051; p=0.835), TAPSE at rest (R=−0.086; p=0.727 and R=0.001; p=0.407) and RV fractional area change at rest (R=−0.282; p=0.257 and R=−0.208; p=0.407). Furthermore, no correlation was found for PV gradient at rest and at peak exercise for right atrial area (R=0.084; p=0.739 and R=0.336; p=0.173), RV diastolic area (R=0.253; p=0.331 and R=0.368; p=0.133) and RV systolic area (R=0.292; p=0.239 and R=0.362; p=0.140).


This study showed that patients with mild-to-moderate PV stenosis present with decreased exercise capacity and ventilatory efficiency compared to age and gender matched controls. However, right ventricular performance and right heart morphology were preserved in the studied patients. Lastly, PV gradient increases linearly with increasing flow.

In adult patients with severe PV stenosis, exercise capacity is limited.5 The haemodynamic response to exercise in patients with mild PV stenosis is not well studied. Mild PV stenosis is considered benign, and patients can participate in recreational or competitive sports. Asymptomatic adult patients with moderate PV stenosis are subjected to closer follow-up, but are not limited in moderate recreational sport participation and should be encouraged to do so.8 ,9

In our study, only four patients had moderate PV stenosis. Nevertheless, the studied population presented with a decreased exercise capacity. This is in contrast to Romeih et al10, who documented a preserved exercise capacity in 11 patients with moderate PS.10 Higher resting heart rate, lower heart rate reserve, peak oxygen levels and peak power all point towards a lower physical condition as compared with an age-matched and gender-matched control group. Additionally, ventilatory efficiency, that is, OUES and VE/VCO2 slope, was equally impaired. These are known exercise parameters that are independent from the endpoint of a maximal exercise test, which have shown their value in different heart disease populations.13 ,14 ,15 Furthermore, it has been shown that both exercise capacity and ventilatory efficiency improve after relieve of right ventricular pressure load in patients with PV stenosis, even if only small improvements of valvular gradient are obtained postintervention.16 ,17 An increase in afterload reserve and the associated increase in stroke volume at peak exercise is the most likely underlying mechanism.16 Contradictory to these findings, no significant difference between patients and controls was observed for peak oxygen pulse or for the slope of the oxygen pulse—work load relationship (figure 2). Both TAPSE and RV fractional area change at rest were within the reference values. These data suggest that stroke volume is not influenced by the presence of mild-to-moderate PV stenosis. Because our patients were asymptomatic, differences in afterload reserve in our patients with mild-to-moderate PV stenosis compared with controls might be less pronounced than in the patient population with more severe obstruction studied by others. Otherwise, in adult patients with only mild PV stenosis, diastolic dysfunction of the RV may contribute to a decreased stroke volume.6 However, in this study, severe diastolic dysfunction was not observed as evidenced by normal right atrial dimensions. Possibly, ventilation-perfusion mismatching due to preferential perfusion of certain regions of the lung might cause impaired ventilatory efficiency. Indeed, depending on the morphology of the valvular stenosis, preferential flow to the left or right lung is noticed. Lastly, changes in respiratory function, oxygen delivery or oxygen extraction in the peripheral muscles could contribute as well.

Furthermore, although pressure load of the RV was significantly elevated at peak exercise, no changes in right heart morphology could be seen with increasing maximal pressure load. The pressure load on the RV at peak exercise can be as high as 90 mm Hg, well beyond what is considered to be physiological and could be associated with RV dysfunction or arrhythmic problems in the long term. Patients included in the study were young, and the consequences of life-long pressure increase still have to be established. However, at present, our studied population with mild-to-moderate PV stenosis does not show morphometric changes in the right heart. This is different to what we found in a cohort of patients with pulmonary hypertension where morphometric changes are strongly related with the pressure load on the right heart.18 This might be because the absolute values of pressure load in patients with pulmonary hypertension tend to be higher at rest and during exercise compared with our study sample. Furthermore, in PV stenosis, several factors are thought to contribute to better preservation of right ventricular morphology. The onset, haemodynamic profile and compensatory mechanisms in pulmonary hypertension are very different compared with patients with PV stenosis. In the studied patients, the pressure load is already present to some extent at birth, it is a purely systolic pressure load and neurohormonal or gene and protein synthesis alterations are thought to play a role in the better preservation of right ventricular morphology.19 ,20 ,21 To what extent each contributes to this favourable remodelling of the RV has still to be investigated.

In summary, the studied patients presented with a decreased exercise capacity, but were able to increase their stroke volume similarly to healthy controls. In these patients, the lower exercise capacity is therefore mainly attributable to a lower physical fitness, and probably not directly related to the higher right ventricular afterload. The preservation of right heart morphology and function is reassuring and confirms the current practice not withholding recreational exercise in these patients.8 ,9 Promotion of physical activity seems therefore safe and should be mandatory in routine follow-up. However, long-term effects of this lesion on right heart morphology and function remain uncertain.

Finally, we showed a strong linear relation between PV gradient and cardiac output during exercise. Consequently, the resting PV gradient was predictive for the peak PV gradient, and from the pulmonary resting gradient, the maximum pressure load at peak exercise can be calculated (y=1.669x+13.29). This linear relationship between pressure and flow indicates again a good ventricular performance. On the other hand, this indicates a fairly fixed valve area in patients with the classical form of PV stenosis. The absence of a dynamic component also indicates the absence of infundibular hypertrophy as is seen in patients with severe PV stenosis. Our studied patients with mild-to-moderate PV stenosis do not appear to present with such remodelling. This indicates that if the PV echocardiographic gradient is only mild to moderately elevated, and if the gradient remains stable throughout follow-up, the pressure load is well tolerated and early intervention is not indicated.


We showed that patients with mild-to-moderate PV stenosis do have impaired exercise capacity. In PV stenosis, the valve area is fixed throughout the exercise. At present, we do not have a reason to exclude patients from recreational sports, as no signs of functional or morphological changes of the right heart were observed in the studied range of valve gradients.


Our results are based on a fairly small study sample, and the study has been done in a single centre. However, we were able to match the patient population with a well-established control population. This made useful comparative statistics possible. The number of patients is relatively low for calculating correlations. However, the group of patients is quite homogenous with no outliers. Therefore, we believe that an absence of correlation can be interpreted, however, with careful consideration. Long-term follow-up data are needed to confirm this.

Key messages

  • What is already known on this subject?

  • Adult patients with mild-to-moderate pulmonary valve stenosis are often asymptomatic and have a good prognosis. Consequently, patients are encouraged to participate in recreational sports. However, the impact of the increased pressure load during exercise on right ventricular morphology and function has not been studied before.

  • What this study adds?

  • We provide a comprehensive morphological and functional evaluation of patients with mild-to-moderate pulmonary valve stenosis during exercise. Although functional capacity was decreased compared with healthy controls, right heart morphology and right ventricular performance were preserved.

  • How might this impact on clinical practice?

  • These findings provide evidence that encouraging physical exercise in patients with mild-to-moderate pulmonary valve stenosis would probably be safe. It also adds to the evidence that no intervention is indicated if the valve gradient is only mild to moderately elevated.


PDM was supported by a grant of the Agency for innovation by Science and Technology in Flanders (IWT).



  • Contributors All authors contributed to the conception and implementation of the study. All have read and approved the final version of this manuscript.

  • Funding Agency for innovation by Science and Technology in Flanders (IWT). (Grant no. 101484).

  • Competing interests None.

  • Ethics approval Ethics committee UZLeuven.

  • Provenance and peer review Not commissioned; externally peer reviewed.

Linked Articles