Author + information
- Received September 15, 1998
- Revision received April 14, 1999
- Accepted July 19, 1999
- Published online November 1, 1999.
- William T Mahle, MDa,
- Gil Wernovsky, MD, FACCa,
- Nancy D Bridges, MD, FACCa,
- Andrea B Linton, MEda and
- Stephen M Paridon, MD, FACCa,* ()
- ↵*Reprint requests and correspondence: Dr. Stephen M. Paridon, The Children’s Hospital of Philadelphia, 34th & Civic Center Boulevard, Philadelphia, Pennsylvania 19104.
We sought to determine if early ventricular volume unloading improves aerobic capacity in patients with single ventricle Fontan physiology.
Surgical strategies for patients with single ventricle include intermediate staging or early Fontan completion to reduce the adverse affects of prolonged ventricular volume load. The impact of this strategy on exercise performance has not been evaluated.
Retrospectively, we reviewed the exercise stress test results of all preadolescents with single ventricle Fontan physiology. “Volume unloading” was considered to have occurred at the time of bidirectional cavopulmonary anastomosis or at the time of Fontan surgery in those patients who did not undergo intermediate staging. Potential predictors of aerobic capacity were analyzed using multivariate regression.
The patients (n = 46) achieved a mean percentage predicted of maximal oxygen consumption (V̇O2max) of 76.1% ± 21.1%. The mean age at the time of volume unloading was 2.7 ± 2.4 years, and the mean age at testing was 8.7 ± 2 years. Intermediate staging was performed in 16 of 46 patients (35%). In multivariate analysis, younger age at volume unloading was associated with increased aerobic capacity (p = 0.003). Other variables were not predictive. The subgroup of patients who underwent volume unloading before two years of age achieved a mean percentage predicted V̇O2max of 88.6% ± 24.1%.
Preadolescents with single ventricle who undergo volume unloading surgery at an early age demonstrate superior aerobic capacity compared with those whose surgery is delayed until a later age.
Patients with single ventricle who have undergone the Fontan procedure demonstrate impaired aerobic capacity during exercise as measured by maximal oxygen consumption (V̇O2max). Previous studies have demonstrated that this impairment is due to the limited ability to increase cardiac output during exercise (1–3). Abnormalities of pulmonary vascular resistance and chronotropic response, as well as compromised systolic and diastolic ventricular function, have been implicated as possible mechanisms contributing to impaired cardiac output during exercise (1,2,4,5). Ventricular dysfunction results from the volume-loaded state of the single ventricle prior to either the bidirectional cavopulmonary anastomosis (BCPA) or the Fontan procedure (6,7). Recent changes in the management of the single ventricle have included an attempt to reduce the adverse effects of prolonged volume load on ventricular function by performing volume unloading surgery at an earlier age.
Limited hemodynamic data suggest that abolishing chronic volume overload at a younger age improves short-term ventricular performance (7,8). It is not known whether very early volume unloading surgery improves longer-term ventricular function. Furthermore, surgical strategies such as intermediate staging and early Fontan completion alter pulmonary blood flow patterns and may potentially affect pulmonary artery growth and resistance. The overall impact of these strategies on exercise performance is not known. Previous studies of aerobic capacity in patients with Fontan circulation have included very few subjects who underwent volume unloading surgery at a very early age. At our institution volume unloading surgery is frequently performed in children less than two years of age. The purpose of this study is to determine whether the strategy of very early volume unloading improves aerobic capacity in patients with single ventricle.
The study population included all children with Fontan circulation who underwent exercise testing at our institution between January 1, 1989 and February 28, 1998. Patients were included in the study if they met the following criteria:
1. they had undergone an intraatrial lateral tunnel-type of Fontan at our institution,
2. they had a baseline arterial saturation of ≥85% as measured by pulse oximetry,
3. they were in sinus rhythm during peak exercise and had not undergone pacemaker implantation, and
4. they were less than 13 years of age at the time of the study.
The study was limited to children less than 13 years of age in order to eliminate the confounding affects of pubertal changes on exercise performance. If patients completed more than one exercise stress test in this period, data from the most recent study was included in the analysis.
We reviewed medical records of all patients in order to obtain information about variables that might be predictive of exercise performance. These variables are listed in Table 1. For the purposes of this study, “volume unloading” surgery was said to have occurred either at the time of the BCPA or at the time of the Fontan, if no intermediate surgery was performed. Ventricular morphology was classified as either single right ventricle (RV) or single left ventricle (LV). In those patients who had undergone fenestration of the Fontan baffle, echocardiograms obtained near the time of exercise testing were reviewed in order to determine if the fenestrations were still patent.
Prior to exercise testing, all patients underwent testing of resting pulmonary mechanics, which consisted of inspiratory and expiratory flow volume loop measurements. Forced expiratory volume at 1 s (FEV1) and functional vital capacity (FVC) were recorded. Breathing reserve at maximal exercise (BR) was calculated and expressed as a percentage, BR = [1−(V̇E/MVV) × 100)], where V̇E is minute ventilation and MVV is maximum voluntary ventilation. As MVV can be difficult to measure in younger children, we calculated an estimated MVV by the equation: MVV = 40 × FEV1. The results of the spirometry were compared with data for healthy children as previously reported by Polgar (9).
Exercise testing was performed on a cycle ergometer (Medical Graphics CPE 2000 or Bosch 601). The cycle testing consisted of a ramp protocol with a 3 min warm-up period with no load and then a continuously increasing work load designed to achieve the subjects maximal work-rate by 10 min of exercise. Subjects who were too short for the cycle (height <125 cm) were tested on a Treadmill (Marquette 1800). These subjects underwent a 1 min incremental protocol with increasing speed and grade. Patients were uniformly encouraged to exercise to exhaustion. Oxygen saturation was measured continuously by ear oximetry.
Metabolic data obtained during testing included minute oxygen consumption (V̇O2), minute carbon dioxide production (V̇CO2), V̇E and the respiratory exchange ratio (RER). These variables were monitored continuously on a breath-by-breath basis using a Sensormedics (Yorba Linda, California) metabolic cart and a mass spectrometer (Perkin-Elmer, Pomona, California).
A 12-lead ECG (Marquette case-12, Milwaukee, Wisconsin) was obtained in all patients prior to testing in the supine and standing positions. Electrocardiogram (ECG) leads II, AVF and V5were displayed continuously on a color monitor. A 12-lead ECG was then obtained each min during the course of exercise testing and during the first 5 min of the recovery period.
Continuous variables were summarized as range and mean value ± standard deviation (SD). The data were analyzed with uni- and multivariate regression. The primary outcome measure was maximum V̇O2expressed as a percentage predicted. Potential predictors of aerobic capacity were evaluated for their association with maximum V̇O2. These independent variables were considered candidate predictors in the multivariate model if the corresponding p value was <0.10. Variables remained in the model if the p value was <0.05. We also divided our patient population into two groups based on age of volume unloading, <2 years versus >2 years, since the exercise performance of the former group has not been well described in the literature. Analysis of outcome variables between these two groups was performed with Student ttest. Statistical analysis was performed with STATA for Windows 95 (STATA Corporation, College Station, Texas).
We identified 46 subjects who met eligibility criteria. The anatomical diagnoses are summarized in Table 2. Heterotaxy was present in three patients, of these, two had RV morphology and one had LV morphology. Males accounted for 72% of the study population. The study patients underwent volume unloading surgery between October 1985 and July 1992, and the mean age of volume unloading was 2.7 ± 2.4 years, range: 2 months to 9.6 years. The Fontan procedure was performed without intermediate staging in 30 patients (65%) at a mean age of 3.5 ± 2.2 years. There were 16 patients (35%) who underwent intermediate staging at a mean age of 36.6 (±24) months, range: 2 months to 7.9 years. For these 16 patients the mean interval from BCPA to subsequent modified Fontan operation was 1.23 (± 0.75) years. Exercise testing was carried out at a mean age of 9.2 ± 2.1 years, range: 5.9 to 12.7 years of age. None of the patients had additional sources of pulmonary blood flow such as anterograde flow from the ventricle.
All patients underwent a lateral tunnel-type Fontan; however, there was some variation in the creation of fenestration of the baffle and exclusion of hepatic veins. Fenestration of the Fontan baffle was performed in 10 patients (22%). In 7 of these 10 patients, the fenestration was presumed to be closed or hemodynamically insignificant since the oxygenation saturation was ≥94% throughout the study. In the three other patients who underwent a fenestrated Fontan procedure, oxygen saturations with exercise were in the range of 90% to 93%. Review of echocardiographic studies done near the time of exercise testing demonstrated that one subject had evidence of a 3 mm fenestration at the time of the exercise test. The two other subjects had no evidence of a patent fenestration by either two-dimensional or color-Doppler imaging. Partial hepatic vein exclusion was performed in six patients. All six of these patients had interventions to abolish this source of right to left shunting prior to the time of exercise testing; three patients had surgical reinclusion of the hepatic veins and three underwent transcatheter closure. There were three subjects who demonstrated an oxygen saturation of less than 90% while exercising. One of these patients had evidence of a small decompressing vein from the superior vena cava to the right upper pulmonary vein noted at subsequent catheterization. In another patient cardiac catheterization demonstrated a small leak in the Fontan baffle. In the third patient no obvious source of right to left shunting could be identified. At the time of testing, 22% of patients were taking ACE inhibitors on a regular basis.
Thirty-six of 46 subjects were tested on the cycle ergometer. The remaining 10 patients were tested on a treadmill. Table 3summarizes the exercise data for all 46 patients.
Maximum oxygen consumption
The V̇O2max ranged from 17.9 to 45.9 ml/min/kg. The mean V̇O2max was 34.5 ml/min/kg. When expressed as a percentage predicted, the range for V̇O2max was 32% to 116%, with a mean value of 76.1%. When independent variables were tested for their association with percentage predicted V̇O2max, volume unloading at a younger age was significantly associated with increased aerobic capacity (Table 4). Figure 1demonstrates the association between age of volume unloading and V̇O2max. The mean V̇O2max in the subgroup of patients who underwent volume unloading at less than two years of age was 88.6% predicted versus 62.5% predicted for those who underwent volume unloading greater than two years of age (Table 5).
The peak RER in our study population was 1.05 ± 0.10. While this is relatively low compared with published adult norms, the peak RER in normal pediatric subjects has been reported to be between 0.98 and 1.11 (10). The peak RER in our study population is also comparable with that described in previous evaluations of exercise performance of patients with Fontan physiology with RER measurements, ranging between 1.01 and 1.07 (11–13).
Heart rate response
The mean resting heart rate for the 46 subjects was 84.1 ± 13.6 beats/min. The mean maximum heart rate (HRmax) was 155.7 ± 21.0 beats/min. When expressed as a percentage of predicted, the mean HRmax was 77 ± 11.4%. None of the candidate variables were predictive of HRmax. The mean HRmax for those patients who underwent volume unloading before two years of age was not statistically different from those who were volume unloaded at greater than two years of age.
There were seven subjects (16%) who did not satisfactorily complete a portion or all of the pulmonary testing. For the remaining 39 subjects, the mean values for both FEV1and FVC were less than 80% predicted. The mean breathing reserve was 19.2 ± 8.1%, the reported normal range for the pediatric population being 20% to 50% (14). Regression analysis failed to show any significant correlation between measurements of pulmonary function and aerobic capacity. Similarly, there is no relation between achieved maximal heart rate and resting spirometry function. Age of volume unloading did not predict pulmonary function in this population.
All patients were in sinus rhythm at peak exercise. One patient had first degree atrioventricular block. Four patients had junctional rhythm at rest. Frequent premature ventricular contractions occurred in two patients. Frequent premature atrial complexes were noted in three subjects.
This study demonstrates that performing volume unloading surgery at a very early age is associated with significantly increased aerobic capacity in patients with single ventricle Fontan physiology. In the subgroup of patients who underwent volume unloading surgery before two years of age, the V̇O2max was 88.6% of predicted. This represents a marked improvement over the previously reported V̇O2max for patients with single ventricle and Fontan circulation.
There are several studies that have documented significant impairment in exercise performance in patients with single ventricle and Fontan physiology. The aerobic capacity in these patients has been reported to be between 48% to 64% of predicted values (4,11,12,15,16). A variety of factors have been shown to affect the aerobic capacity in patients who have undergone the Fontan procedure, these include: direct atriopulmonary anastomosis, nonconfluent pulmonary arteries preoperatively, older age at testing and female gender (4,12,17). Age at operation has not been shown to predict exercise performance. However, the populations described in these previous series included very few patients who underwent volume-unloading surgery (either BCPA or Fontan) at a very young age. In the largest study of exercise performance in patients with Fontan physiology reported by Durongpisitkul et al. (12), the mean age of Fontan operation was 11.8 ± 8.6 years, and the earliest that the Fontan procedure was performed was at two years of age.
Early volume unloading
Young age has been reported as a risk factor for successful completion of the Fontan procedure (18–20). However, more recent reports suggest that the Fontan operation can be performed at a younger age with an acceptable morbidity and mortality (21,22). At our institution the combined mortality for intermediate staging and completion of the Fontan procedure is <1% (23). The impetus for performing the volume unloading surgery at a younger age is the known risk of prolonged volume load on the single ventricle. Utilizing catheterization data, Uemura et al. (8) have shown that younger age at the time of the Fontan is associated with improved ejection fraction in the postoperative period. Analysis of wall stress has demonstrated that there is a reversal in abnormal contractile mechanics in the single ventricle after the Fontan procedure (7). However, this capacity for recovery diminishes with age at surgery.
At our institution, changes in surgical management of patients with single ventricle have included an effort to perform the Fontan operation at a younger age in hopes of reducing the adverse affects of prolonged ventricular volume load. Currently, most patients undergo staged palliation by performing an intermediate procedure such as a BCPA. The BCPA results in a significant diminution of the ventricular volume (24,25). In patients who undergo staged palliation, the BCPA appears to be the major volume unloading procedure (24). The BCPA is typically performed in the first year of life. Between 1990 and 1995 the mean age for the BCPA at our institution was 7.3 months (26). It has not been possible to evaluate the effects of very early ventricular volume unloading previously, as the first cohort of patients have only recently reached an age at which valid exercise testing can be performed.
Besides diminished aerobic capacity, patients with Fontan physiology have a blunted heart rate response to exercise. Maximal heart rate has been reported between 70.0% to 89.9% of the predicted HRmax (4,12,15). Sinus node dysfunction may be due to the atriotomy performed at the time of the hemi-Fontan and Fontan procedures. Since all subjects in this study had a lateral tunnel-type Fontan, the risk for impaired heart rate response is significant. The mean HRmax in our study was 157.5 beats/min which represents 76.9% of that predicted and is comparable to the chronotropic response noted in other studies of patients with Fontan physiology. Multivariate analysis did not identify any variables predictive of HRmax. In particular, early volume unloading was not associated with improved heart rate response during exercise.
The data from this study suggest, therefore, that the improved oxygen consumption associated with early volume unloading is due to an increase in stroke volume, rather than conserved chronotropic response. This would concur with previous studies that have shown that the major determinant of aerobic capacity in patients with single ventricle is the ability to increase stroke volume. Gewillig et al. (6) stratified 42 patients with Fontan physiology based on exercise performance and found that the 10 patients with the highest oxygen consumption at peak exercise had a higher stroke volume than the 10 patients with the lowest oxygen consumption. There was no difference between the two groups, however, with respect to chronotropic response. Cortes et al. (1)performed echocardiography during exercise testing of patients with Fontan circulation and found an inability to increase stroke volume near maximal exercise. Data from Fukushima et al. (27)have demonstrated that there is an association between elevated end-diastolic pressure at peak exercise and lower V̇O2max in patients with single ventricle. This would support the hypothesis that prolonged volume load adversely affects ventricular function by impairing filling of the ventricle during exercise, which in turn limits stroke volume.
Along with impairment of V̇O2max and heart rate response to exercise, Fontan patients have alterations in pulmonary mechanics. Previous thoracotomies may be one potential cause. Additionally, significant lung perfusion abnormalities have been noted in Fontan patients (14,28). There is a theoretical concern that pulmonary artery growth may be adversely affected by early conversion to nonpulsatile blood flow. This has been one argument against very early BCPA or Fontan surgery. In our population all the measured indexes of pulmonary function were low, but there was no significant difference between the early and late volume unloading groups.
The major limitation of this study is the retrospective study design with a nonselected patient population. There is potential for selection bias with respect to both the timing of volume unloading surgery and elective referral for exercise testing. There are many factors that influence the timing of volume unloading surgery, including significant variation in the management of patients with single ventricle by individual cardiologists and dramatic changes in management philosophy at the time these patients were undergoing palliative surgery. It is possible that volume unloading surgery might be delayed in patients with poorer ventricular performance. However, the approach at our institution is to attempt volume unloading at a younger age in those patients with poor hemodynamics in order to reduce the volume load on an already compromised ventricle. Such a strategy would presumably create a bias against those patients who underwent the BCPA or Fontan procedure at a younger age.
There is an additional potential for selection bias in our study population since only a fraction of those patients who underwent the Fontan procedure completed exercise testing. As is true of most reports of exercise performance, those patients with very poor hemodynamics are unable to perform the exercise testing and are, therefore, excluded. Furthermore, neurologic impairment is not uncommon in this population and many patients cannot be evaluated for this reason. While many physicians consider exercise testing routine in the management of school-age patients with single ventricle, others do not. However, when one analyzes the subset of our patients who underwent volume unloading at age two years and above, this group is comparable to populations with Fontan physiology described previously in the literature in terms of aerobic capacity. This would suggest that the overall improvement in aerobic capacity for our entire group is due to the subgroup of patients who underwent early volume unloading.
Conclusions and clinical implications
The findings of this study demonstrate that early ventricular volume unloading is associated with improved aerobic capacity. This appears to be due to improved ventricular function rather than preserved chronotropic response. The data from our study, however, do not suggest an optimal age for performing volume unloading surgery. For example, is it more beneficial to perform the BCPA at four months of age versus nine months of age? Since the number of patients who underwent BCPA was small, we also cannot determine whether the interval between BCPA and Fontan completion impacts aerobic capacity in these patients.
It remains to be seen whether the improved aerobic capacity in this prepubertal population will persist as these patients age. Additionally, it will be important to determine what impact early ventricular volume unloading has on other issues relevant to patients with Fontan physiology such as the incidence of arrhythmias and protein losing enteropathy, as well as the potential effects on neurodevelopmental outcome. Accordingly, serial follow-up including assessment of exercise performance, especially in that subgroup of patients who are managed with intermediate staging, will be important in optimizing surgical management.
This study was presented in part at the 71st Scientific Sessions of the American Heart Association, Dallas, Texas, 1998.
- bidirectional cavopulmonary anastomosis
- breathing reserve at maximal exercise
- forced expiratory volume at 1 second
- functional vital capacity
- maximum heart rate
- left ventricle
- maximum voluntary ventilation
- respiratory exchange ratio
- right ventricle
- standard deviation
- minute carbon dioxide production
- minute ventilation
- minute oxygen consumption
- maximal oxygen consumption
- Received September 15, 1998.
- Revision received April 14, 1999.
- Accepted July 19, 1999.
- American College of Cardiology
- Gewillig M.H,
- Lundstrom U.R,
- Bull C,
- Wyse R.K,
- Deanfield J.E
- Ben Shachar G,
- Fuhrman B.P,
- Wang Y,
- Lucas R.V Jr.,
- Lock J.E
- Driscoll D.J,
- Danielson G.K,
- Puga F.J,
- Schaff H.V,
- Heise C.T,
- Staats B.A
- Troutman W.B,
- Barstow T.J,
- Galindo A.J,
- Cooper D.M
- Sluysmans T,
- Sanders S.P,
- van der Velde M,
- et al.
- Polgar G,
- Promadhat V
- Rowland T
- Durongpisitkul K,
- Driscoll D.J,
- Mahoney D.W,
- et al.
- Matsushita T,
- Matsuda H,
- Ogawa M,
- et al.
- Rhodes J,
- Garafano R.P,
- Bowman F.O,
- Grant G.P,
- Bierman F.Z,
- Gersony W.M
- Harrison D.A,
- Liu P,
- Walters J.E,
- et al.
- Rosenthal M,
- Bush A,
- Deanfield J,
- Redington A
- Fontan F,
- Baudet E
- Bartmus D.A,
- Driscoll D.J,
- Offord K.P,
- et al.
- Pearl J.M,
- Laks H,
- Drinkwater D.C,
- Capounya E.R,
- George B.L,
- Williams R.G
- Weber H.S,
- Gleason M.M,
- Myers J.L,
- Waldhausen J.A,
- Cyran S.E,
- Baylen B.G
- Berman N.B,
- Kimball T.R
- Forbes T.S,
- Gajarski R,
- Johnson G.C,
- et al.
- Cohen M.I,
- Wernovsky G,
- Vetter V.L,
- et al.
- Fukushima N,
- Matsuda H,
- Ichikawa H,
- et al.