In this article, the authors describe their use of the Shape-HF system in patients with cardiovascular disease. They also discuss other applications of the technology, such as in cardiac resynchronization therapy and pulmonary arterial hypertension. Background Currently, chronic cardiovascular disease — specifically, congestive heart failure (CHF) — represents one of the biggest and most expensive disease processes that face our healthcare system. While research has identified several therapies known to improve morbidity and mortality in heart failure (i.e., beta blockers, ACE inhibitors [ACEIs] and device therapies), objective assessments of overall response to therapy and prognosis remain lacking. At this time, the tools available to the physician that can help estimate prognosis include the New York Heart Association (NYHA) functional classification, the six-minute walk test, and cardiopulmonary exercise testing (CPET). The difficulties associated with the first two tools are that they introduce interobserver variability as well as issues related to patient effort, although both have been correlated with prognosis in CHF. While the first two assessment tools can be impacted by a number of conditions, the usage of CPET does offer a more objective measure of patient functional capacity. As data has already shown, the peak oxygen consumption (VO2) obtained during CPET has relevance both in patient prognosis as well as determining when patients should be referred for cardiac transplant.1,4,11 This measure has also been useful in determining response to various therapeutic interventions in the CHF population.1 While this data grew rapidly in the early 1990s, the identification of ventilatory parameters did not start to be recognized until the later 1990s.1 The measure of ventilatory efficiency (VE/VCO2) started to gain more attention when certain limitations in measurement of VO2 became evident.1,4,11 In particular, some studies have suggested that measurements of peak VO2 were not reflective of the addition of beta blockers, whereas the VE/VCO2 slope retains its prognostic significance despite medical therapy for CHF.1,11 In addition, the measurement of peak VO2 has required patients to achieve maximal exertional capacity, which is difficult given the nature of the disease processes being investigated. The Shape System At the University of Illinois at Chicago (UIC), we have been able to obtain early experience with a submaximal cardiopulmonary exercise system (Shape Medical Systems, Inc., St. Paul, MN) (Figure 1). While the data supporting the usage of cardiopulmonary testing at maximal exercise are robust, there are limitations to its usage as previously identified. Measures such as VE/VCO2, oxygen pulse (VO2/HR), oxygen uptake efficiency (VO2/log VE), and partial pressure of end-tidal carbon dioxide (ETCO2), have already been shown to carry significant prognostic value.1,2 Additionally, other measures such as heart rate recovery and chronotropic response index provide additional useful information. Combining these measures, already clinically proven in prior studies of CHF, along with the innovation of measurement at submaximal exercise, we believe make the Shape system particularly appealing for clinical use. The test takes 15 minutes and involves measuring ventilation parameters while the patient exercises on a treadmill at a very low intensity of one mile per hour with the treadmill set at a 2% grade. The device itself includes five components: a data analyzer, disposable patient interface or mask, a pulse oximeter, a computer and a printer. As the patient exercises at a steady state heart rate, the physician adjusts therapy settings every two minutes, enough time for the adjustments to be reflected in breathing physiology. At the end of the test, during which four to five therapy settings are tested, the Shape-HF system uses a proprietary computer algorithm to rank the physiological response to exercise at each setting. The physician then reviews the results and chooses the therapy setting he or she believes is most appropriate for the patient. The data supporting the use of ventilatory efficiency as a prognostic marker superior to that of peak VO2 comes from numerous studies comparing the two measures.1,11 As these studies indicate, the rise in VE/VCO2 slope that accompanies worsening CHF is likely attributable to increased dead space ventilation as well as decreased perfusion compared to the physiologic demands of the body.1 More importantly, this finding has been confirmed across a wide range of patient groups, which include both systolic and diastolic heart failure as well as patients on medical therapy with agents such as beta blockers and ACEIs.1 Perhaps the most important element of measuring ventilatory efficiency comes from the fact that it is not dependent on achieving peak exercise. To date, there have been numerous studies that have compared the prognostic significance of submaximal exercise testing with VE/VCO2 measurement compared to maximal exercise and measurement of peak VO2. These studies have shown that evaluation of ventilatory efficiency at submaximal exercise consistently demonstrates retained prognostic power over that of peak VO2.1,12 The identification of this relationship highlights one of the strongest features of Shape technology — that patients with chronic cardiovascular disease do not need to achieve peak exercise to obtain meaningful data. In addition to the ventilatory measures discussed, the heart rate indices that are obtained via Shape testing also add a significant amount of information to treatment of chronic cardiac diseases (Figure 2). The two measurements that the Shape system specifically evaluate are heart rate recovery and chronotropic response index. Extensive data has already indicated that chronotropic response to exercise can identify patients at risk for increased mortality. Specifically, the inability to augment the heart rate in response to exercise is associated with worse outcomes. This is a relationship that continues to apply despite medical therapy. Similar to the augmentation seen with exercise, the heart rate recovery after exercise is also a powerful measure. Data exist that indicate a slower heart rate recovery can be associated with worse overall prognosis, as well as an increased risk of death, depending on the severity of the response. We believe the Shape-HF system represents an important leap forward in the evaluation of patients with CHF. By having patients perform submaximal exercise and collecting data across a variety of ventilatory and chronotropic indices, it has the ability to allow tailored therapy to the individual patient. In addition, the ability to provide real-time data interpretation using the Shape algorithm makes this a tool that more practitioners can take advantage of in the office setting. Evaluation in Congestive Heart Failure To date, there is a great deal of information using ventilatory parameters and correlating these values with prognosis in CHF. While aerobic capacity and peak VO2 had been looked at extensively, the use of ventilatory efficiency appears to retain prognostic significance despite changes in therapy, and across a range of cardiovascular disorders.1 Specifically, the VE/VCO2 slope is still predictive of mortality in the setting of systolic and diastolic heart failure, right-sided heart failure, cardiac transplant patients and in the setting of medical therapy and cardiac resynchronization therapy.1,3 Given the extensive literature already available, the ability to estimate prognosis based upon baseline VE/VCO2 slope has already emerged. Patients with lower baseline values generally experience fewer cardiovascular events.1 However, as the VE/VCO2 slope rises, clinical events, including hospitalization, tend to increase as well. Given the interest in developing risk models and assessing response to therapy, these ventilatory parameters seem to offer a unique marker by which we can follow CHF patients. Application in CRT Cardiac resynchronization therapy (CRT) is a significant therapeutic advance in the treatment of advanced systolic heart failure with evidence of dyssynchrony on resting electrocardiogram. Early trials have demonstrated significant improvement in both morbidity and mortality in patients with class III and IV heart failure, meeting the appropriate criteria for implantation.5,9 Despite evidence-based criteria supporting placement in patients with an ejection fraction of less than 35%, a wide QRS (left bundle branch block pattern), and NYHA functional class III or IV symptoms, there still remains a significant number of non-responders. The literature generally reports that approximately 30% of implanted patients fail to respond clinically to CRT.8 These non-responders continue to have higher rates of hospitalizations, lower quality of life, and increased mortality.8 Currently, it is not clear whether these non-responders represent patients who simply do not show benefit with CRT as opposed to patients who do not respond due to inadequate programming of their CRT devices. Extensive efforts have been made to better select patients for CRT, based on data derived from echocardiography, radionuclide ventriculography, and cardiac computed tomography, and magnetic resonance imaging.8 These studies have failed to identify common variables that reliably predict response to CRT. Early data examining changes in ventilatory parameters in the setting of CRT have already demonstrated significant improvements in ventilatory efficiency as well as heart rate profiles with exercise.3,13 Specifically, when compared to patients without CRT, those undergoing biventricular pacing show a significant improvement in peak VO2 as well as ventilatory efficiency as measured by VE/VCO2.3,13 The magnitude of benefit achieved with CRT seems to be strongest in those patients with the worse ventilatory parameters prior to starting therapy.3,13 Besides the ventilatory parameters, there is also data suggesting improved heart rate response to exercise with CRT, as measured by both baseline heart rate and chronotropic response to exercise.3 These studies, as well as several others, support the positive response to CRT that can be assessed by cardiopulmonary testing. While studies identified patients with worse starting parameters as those most likely to benefit from CRT, the question still remains on how to best optimize the various programmable settings available in CRT following implantation. The two settings that are the most targeted in this situation are the atrioventricular (A-V) delay and the ventriculo-ventricular (V-V) timing interval. There have been numerous efforts made to identify an approach to optimal programming of CRT devices. Studies have focused on assessing echocardiography, intracardiac electrocardiography, radionuclide ventriculography, as well as finger plethysmography.6,7,10 Despite these efforts, there is yet to be a consensus method that consistently yields optimal results regarding A-V or V-V timing. More importantly, these methods can be time consuming, difficult to assess serially, and generally not correlated with measures known to have prognostic impact in the treatment of heart failure. To date, there have been studies examining submaximal exercise testing using the Shape system to help optimize both A-V and V-V timing intervals in CRT devices. Through the use of an algorithm designed to assess the impact of various A-V and V-V intervals during submaximal exercise, the Shape system has reliably identified the timing cycles that correlate with the best ventilatory parameters. The ability to perform testing with various timing cycles, monitor serial performance, as well as correlate the results with measures known to have prognostic impact are all benefits of the Shape system. While these investigations are still in their infancy, the data is promising to date and might represent a new level of tailored therapy to the individual in the use of device therapy for heart failure. Future Directions In addition to CRT, another potential application of the Shape technology and submaximal cardiopulmonary exercise testing lies in the management of pulmonary arterial hypertension (PAH). In recent years, PAH has received increasing attention, with its attendant significant morbidity and mortality. Similar to heart failure, treatment of PAH has been difficult to follow due to the lack of objective measures of response to therapy. Much like CHF, assessments are currently limited to subjective assessment of patient symptoms, functional class, or six-minute walk assessments which can be influenced by patient effort. Due to these factors, the identification of a reproducible and reliable indicator of patient response and prognosis is of particular interest in the pulmonary hypertension community. There have already been studies performed that correlate prognosis in PAH to measurements obtained during cardiopulmonary testing. In particular, peak VO2 has been used to assess severity in patients with primary pulmonary hypertension.14 Although peak VO2 has been validated previously, obtaining peak VO2 has its own limitations as previously identified. Through assessment of submaximal exercise and ventilatory parameters such as VE/VCO2 and ETCO2, it was believed that severity of pulmonary hypertension could be evaluated, and serial assessments performed, that might allow a response to therapy to be determined. Data from Yasunobu et al revealed that with increasing severity of pulmonary hypertension as determined from mean pulmonary artery pressures, the VE/VCO2 value also tended to increase, indicating more severe disease.14 The ventilatory findings in these studies, which also show a significant decrease in the ETCO2 with exercise in PAH, have the potential to offer a unique means of following these patients clinically.14 We believe that by utilizing Shape technology, these patients can be assessed serially at submaximal levels, obtaining measures that directly correlate with disease severity and overall prognosis. Conclusions Existing tools for the assessment of chronic cardiovascular conditions such as CHF and PAH are highly subjective, subject to significant interobserver variation, and vary a great deal with patient effort. While cardiopulmonary testing and peak VO2 have been correlated with both disease severity and prognosis in these patient populations, limitations include the inability to reflect usage of beta blockers and the need to perform maximal exercise to obtain its value. In patients with these significant cardiovascular limitations, the utility of submaximal testing is a closer estimation to routine physical activity and as such is the best level at which to obtain measures of ventilatory efficiency. This hypothesis has been confirmed in studies utilizing measures of VE/VCO2 and correlating this data to overall prognosis. Additionally, the ease of performance of submaximal testing and Shape algorithms make its usefulness in evaluating both response to CRT and optimization of CRT of particular value. Overall, we believe the usefulness of Shape testing and its increased application in various clinical states make it an exciting new application ready for expanded usage.