We examined a consecutive cohort of patients with HCM who were referred to our hospital for Doppler ultrasound during exercise and cardiopulmonary tests. HCM was diagnosed in the presence of nondilated hypertrophic left ventricle (LV) with a maximum thickness of ≤15mm in index cases or >12mm in their family members, and with no other possible cause.16

We performed a complete clinical evaluation of all patients, an at-rest Doppler ultrasound analysis (with tissue Doppler), and a stress test. Two-dimensional and M-mode measurements, as well as the Doppler analysis, were carried out using a Sonos 7500 ultrasound system (Philips Medical Systems; Andover, Massachusetts, United States). Baseline values and those following exercise were taken with the patient in left lateral decubitus, and were stored digitally for review and analysis using the Xcelera system (Philips Medical Systems; Andover, Massachusetts, United States). All measurements were then analyzed by the same individual, blinded to the stress test results. For each measurement, 3 different cardiac cycles were averaged, in accordance with the American Society of Echocardiography guidelines.17

All patients gave informed consent to perform the test and the study protocol was approved by our clinical research ethics committee.

LV wall thickness was measured in the short axes of the LV at the mitral valve and papillary muscles in order to obtain maximum wall thickness. The LV ejection fraction was calculated using the Simpson method. Mitral Doppler readings were obtained from the apical 4-chamber view, placing the volume sample at the edge of the mitral leaflets. Peak E-wave and A-wave velocities were averaged in 3 consecutive cycles. Mitral ring velocities were obtained using tissue Doppler placing the volume sample in the lateral/septal ring. We obtained 3 to 5 cycles and averaged 3 for the early (E’) and late (A’) diastolic waves, as well as for an evaluation of diastolic intervals. In accordance with the classification system by Garcia et al.,18 we established 4 different filling patterns in mitral flow: normal, prolonged isovolumetric relaxation, pseudonormal, and restrictive. The LVOT was measured using color-guided continuous wave Doppler. We considered obstructive entities those that reached a gradient>30mmHg in the stress test. The classification of mitral regurgitation was based on the planar analysis of the regurgitating surface: mild (1-4cm2), moderate (4-8cm2), and severe (>8cm2). Left atrial volume indexed for body surface area was obtained using the 2-plane Simpson method (mL/m2).19

Following the baseline Doppler ultrasound examination, we administered a stress test limited to the symptoms present, using a treadmill following the Bruce protocol. Patients were urged to exercise until exhaustion. Heart rate, blood pressure, and a 12-lead electrocardiogram were monitored and registered at rest, every 2min during exercise, and 5min after exercise. Gas exchange values and ventilation parameters were obtained using a pneumotachograph and CPX System analyzer (Medical Graphics Corp.; St. Paul, Minnesota, United States), calibrated as necessary before each test. The VO2, VCO2, and other ventilatory parameters and calculations were determined during each respiration and averaged every 10s. Before starting the test, the respiratory exchange rate had to be≤0.85. VO2max was established as the highest value of the 10-s averages during the last minute of exercise (in mL/kg/min). Functional capacity was determined as the percentage of the observed maximum VO2max as compared to the theoretical maximum value based on age, sex, and body surface area.20 The criteria for interrupting the test were exhaustion, severe arrhythmia, severe hypertension (systolic>240mmHg or diastolic>110mmHg), hypotensive response (>20mmHg), or limiting symptoms. We defined an abnormal pressure response as an increase<25mmHg or a drop>10mmHg during exercise.21

When any of the interruption criteria for the test was reached, the patient was rapidly placed on a gurney, in the same left lateral decubitus position as before. We then recorded 2 cycles of the apical 4-chamber view, 2-dimensional color images, 5 continuous Doppler LVOT cycles, pulsed mitral Doppler, and tissue Doppler of the lateral/septal ring. All images were taken 1min after peak exercise. We considered pseudonormalization of diastolic flow induced by exercise as the change from a prolonged at-rest relaxation pattern to a pseudonormal or restrictive pattern with exercise (Fig. 1).

Figure 1.

Mitral flow at rest (A) and after exercise (B) in patients with a prolonged isovolumetric relaxation pattern at rest, which becomes pseudonormal during exercise, eith an inverted E/A ratio.

(0.13MB).

We used the Kolmogorov-Smirnov test to determine whether each variable exhibited a normal distribution. Variables with a normal distribution were expressed as mean (standard deviation), and those without a normal distribution were expressed as a median [interquartile range]. Qualitative variables were expressed as percentages. The differences between groups were compared using chi-square, Mann-Whitney U, Student t tests or analyses of variance, as appropriate. To evaluate the relationship between variables and VO2 we used Pearson's and Spearman's correlation coefficients. We used a backwards multiple regression analysis to determine which independent factors were associated with functional capacity, expressed as percentage of VO2max. We introduced variables with a significant association resulting from the simple analysis (P<.05) into the model. We used two different models: one with echocardiographic and at-rest Doppler values and another including all variables for both at-rest and exercise tests. In addition to analyzing the overall patient group, we assessed patients with and without obstruction separately to determine the independent predictors for one presentation of the disease or another. We considered a P-value<.05 as statistically significant. For the statistical analysis, we used SPSS statistical software, version 13 (SPSS, Inc.).

In total, we registered 103 patients referred to our department for stress tests including VO2 and Doppler ultrasound during exercise; 16 were excluded due to persistent atrial fibrillation. Therefore, the study included 87 patients (29 female; mean age 51.1 [16.4] years) (Table 1). The majority of patients (43.9%) had functional class II according to the New York Heart Association scale; no patients had undergone operations or alcohol septal ablation previously, and 9 patients carried pacemakers.

Patients walked for a mean 7.15 (2.4) min, the maximum heart rate was 130.1 (25.8) bpm, and the respiratory exchange rate reached 1.01 (0.08). There were no observable complications. One patient suffered a paroxysmal atrial fibrillation at peak exercise, with spontaneous reversion during the first minute of recovery. An abnormal pressure response was observed in 23 patients, although only one of them was hypotensive. Ten of these patients were under treatment (7 with beta blockers and 3 with calcium antagonists).

The results from the ultrasound stress tests are summarized in Table 2. We did not observe segmental abnormalities in the induced contractility during exercise. The LV ejection fraction increased from 65.9 (9.1)% (range: 53%-85%) to 71.5 (9.1)% (range: 57%-85%) (P<.001).

In general, mitral and ring flow velocities increased with exercise (Table 2). There were significant increases in E-wave and A-wave values, but the E/A ratio was constant. We also observed a proportional increase in ring E’ wave values with no differences in the E/E’ relationship after exercise as compared to baseline values.

In some patients, we observed that exercise deteriorated mitral filling patterns (Table 2). At rest, 14 of the 41 patients with a prolonged isovolumetric relaxation pattern developed a pseudonormal or restrictive pattern with exercise (Fig. 1). This pseudonormalization was associated with a greater left atrial volume indexed for body surface area (47.3 [14.6] mL/m2 vs 32.4 [14.5] mL/m2; P<.05), worse functional capacity (66.5 [11.4]% vs 81.8 [20.8]%; P<.05), higher E/E’ ratio following exercise (19.4 [7.9] vs 12.5 [6.4]; P=.015) and more frequent mitral regurgitation following exercise (71.4% vs 25.9%; P<.01).

At rest, 23 patients experienced obstruction (31.5%), and another 20 patients developed gradients>30mmHg with exercise. These 43 patients compose the group with obstructive patterns. Patients with obstruction had more dilated left atria and a greater wall thickness (Table 3). Patients with obstructive patterns had a higher mitral regurgitation value and E/E’ ratio at rest and after exercise. The VO2max (in mL/kg/min or as functional capacity) of patients with obstruction was significantly lower than in nonobstructed cases (Table 3).

Table 4 shows the correlations between the different variables and VO2max. An obstruction at the peak of exercise>30mmHg was correlated with VO2 (r=−0.384; P<.001). However, we did not observe a significant relationship between VO2 and maximum gradient at rest or at exercise. Age (r=−0.28; P=.001), sex (r=−0.52; P<.0001) and left ventricular and atrial volume were correlated with VO2. As regards Doppler parameters, the lateral/septal E/E’ ratios and time between mitral E wave and Q-wave (QE interval) were significantly correlated with VO2, both at rest and upon exercising (Fig. 2). An increase in the septal ring E’ wave (r=0.396; P=.001) was associated with a higher VO2.

Figure 2.

Simple correlation between oxygen consumption and septal E/E’ ratio at rest (A) and after exercise (B). VO2max, maximum oxygen consumption.

(0.17MB).

In the multiple regression analysis, the model including at-rest variables yielded an adjusted R2 of 0.379, and the only factors independently associated with functional capacity were sex (P=.001) and left atrial volume (P=.001). The model that included both at rest and exercise variables showed an adjusted R2 of 0.49. In this model, the independent factors were age, left atrial volume index, obstruction during exercise, and E/E’ and E/A ratios after exercise (Table 4). In obstructive cases, these same variables also were independently associated with functional capacity, with a predictive value>50% (adjusted R2=0.53). In nonobstructed cases, the determining factors for functional capacity were left atrial and ventricular volumes (adjusted R2=0.39).

The analysis of the lateral ring was included halfway through the study, which is the reason why this variable was assessed in only 36 of the 87 patients. In the analysis of these 36 patients we did not observe that variables obtained from the lateral ring measurement improved compared to those of the septal ring, and it is not possible to conclude whether larger samples could improve the predictive value of these variables for functional capacity.