With the advent of implantable hemodynamic monitoring for patients with heart failure (HF), a direct relationship between elevated pulmonary artery (PA) pressures and risk of HF hospitalization is now established.1 Furthermore, reduction in filling pressures relates to a reduced risk of HF hospitalization.1–4 Hemodynamic-guided HF management leads to higher frequency of increases and decreases in diuretic dosing,5 increases in neurohormonal antagonist dosing, 2,5 and reduction in cardiac filling pressures,2 which is directly associated with lower HF hospitalization risk.1–4 These effects are seen in patients with reduced ejection fraction (EF), as well as those with preserved EF.6 Furthermore, PA pressure–guided HF management reduces 30 day readmissions following discharge after an index hospitalization.7 These data were reported from the CardioMEMS Heart sensor Allows for Monitoirng of Pressures to Improve Outcomes in NYHA Class III heart failure patients (CHAMPION) Trial, which evaluated hemodynamic-guided HF management in 550 patients using an implantable sensor system (CardioMEMS HF System; St. Jude Medical, Atlanta, GA) and a randomized follow-up time averaging 18 months. The trial enrolled previously hospitalized subjects determined to be New York Heart Association (NYHA) Class III and ACC Stage C at the time of sensor implantation.8 Long-term clinical benefits of remote hemodynamic monitoring are consistent across all hemodynamic monitoring trials9 leading to the recommendation of this clinical management strategy in the 2016 European Society of Cardiology Heart Failure Management Guidelines.10
Clinical disease progression in patients with HF, however, can still occur despite maximal guideline-directed medical therapies (GDMTs) and hemodynamic-guided care. Recognition of this progression is important because timing of advanced HF therapies, namely left ventricular assist device (LVAD) and transplants, continues to be a clinical problem impacting overall outcomes. Clearly, mechanical circulatory support is superior to medical management in advanced HF patients who progress to ACC/AHA Stage D refractory NYHA Class IV HF, and earlier intervention is associated with better long-term results.11,12
In this report, we summarize the characteristics of patients in the CHAMPION Trial who developed worsening HF despite best medical therapy during the follow-up period and deteriorated to the point of requiring LVAD implantation. A previous study developed the hypothesis that serial monitoring of cardiac filling pressures might be useful in the management of LVAD patients.13 Although multiple clinical variables are involved in the decision to provide LVAD support, implantation of a wireless hemodynamic monitoring system in advanced HF patients may provide physicians with physiologic information needed to improve the timing of LVAD implantation. In addition, physician knowledge of hemodynamic information may allow for more dynamic management of the LVAD device settings after implantation and may improve LV unloading when compared with current methods.
The CHAMPION Trial enrolled 550 NYHA Class III patients, who had been hospitalized for HF in the previous year, at 64 centers in the United States and is described in detail elsewhere.2–7 Patient enrollment occurred from September 2007 to October 2009. Patients were enrolled regardless of left ventricular EF or HF etiology and received all appropriate GDMT at optimal or best-tolerated stable doses.10,14 Major exclusion criteria included a history of recurrent pulmonary embolism or deep venous thrombosis, cardiac resynchronization therapy device implantation within the preceding 3 months, and Stage IV or V chronic kidney disease (GFR < 25 ml/min). The study complied with the Declaration of Helsinki, the institutional review board of each participating center approved the study protocol, and all patients provided written informed consent.
Wireless Implantable Hemodynamic Monitoring System Description
The wireless implantable hemodynamic monitoring system (CardioMEMS HF System) uses a wireless microelectromechanical (MEMS) sensor, which does not require batteries or leads. The PA sensor is part of a system that includes a patient interrogation device for home uploads of PA pressures and a secured website to display hemodynamic information (Merlin.net, St. Jude Medical) (Figure 1) and is described elsewhere in detail.2,4–8 An important characteristic of the system germane to this study is that the sensor is implanted in a distal branch of the PA and remains functional post LVAD implantation and heart transplantation.
In the CHAMPION Trial, all patients were implanted with the PA sensor and then were randomly assigned to either a treatment group or a control group.8 Treatment group patients were managed using remotely obtained PA pressures in addition to standard of care HF monitoring of clinical signs, symptoms, and weights. Control patients, whose hemodynamic data from the PA sensor were uploaded daily, but not available to the physician during the randomized period, were managed according to standard HF management only. All patients regardless of randomization were instructed to take daily pressure readings from home using the system. The primary efficacy end-point of the CHAMPION Trial was the rate of HF hospitalizations between groups and was assessed at 6 months of follow-up. However, each patient remained in their original study assignment until the last patient completed 6 months of follow-up. This allowed an efficacy evaluation after an average randomized follow-up time of 18 months.4 For the current analysis of patients requiring advanced therapy during the randomized follow-up, two time points of interest were chosen: time from randomization to LVAD implant and time from LVAD implant to either transplant or the conclusion of the primary follow-up period.
The CHAMPION Trial protocol censured patients at the time of advanced therapy (LVAD or transplant), and any hospitalization events after the advanced therapy were captured, but not adjudicated or counted toward the primary end-point. Furthermore, although patients remained in their originally randomized groups, remote hemodynamic monitoring was optional after ventricular assist device (VAD) or transplant. Direct interrogation of the sensor during a hospitalization was not allowed by protocol to ensure continuity of patient masking of their randomized group. As a result, both direct and remote hemodynamic monitoring after VAD or transplant were inconsistent and preclude systematic analysis of outcomes.
All of the analyses described in this report are post hoc. Group statistics are presented as mean ± standard deviation (SD) unless otherwise indicated. Clinical profile data were analyzed between groups using a t-test for independent samples when comparing continuous data and a Fisher’s exact test when comparing proportions. Changes within groups were analyzed using a t-test for paired samples when comparing continuous data and a McNemar test when comparing proportions. All pressure readings obtained from the PA monitoring system were used in the pressure analysis. Proportions of patients bridging to transplant were analyzed using a Fisher’s exact test. Time to event analyses was performed using Kaplan–Meier methods and a log-rank test. Interagency Registry for Mechanically Assisted Circulatory Support (INTERMACS) profiling was performed as specified in the published guidelines.15,16 Significance was set at the 0.05 level, and all analyses presented are by intention to treat.
The total follow-up time for the primary randomized study period averaged 18 months for patients in the CHAMPION Trial. During this randomized period, there were a total of 27 patients (15 treatment and 12 control patients) who underwent LVAD implantation. As shown in Table 1, the LVAD cohort had significantly higher creatinine levels, higher PA pressures, lower systemic pressures, and no differences in cardiac output compared with the rest of the CHAMPION population at the time of PA pressure sensor implantation. More patients in the LVAD cohort were prescribed mineralocorticoid antagonism (Table 1). Among the 27 patients in the LVAD cohort, the treatment and control groups clinical profiles were similar at the time of PA sensor implant (Table 2).
LVAD implantation occurred at an average of 211.7 ± 174.6 days after randomization (range 20–607 days) for the entire patient cohort. At the time of LVAD implantation, the clinical profiles between the treatment and control groups were similar (Table 3). However, when comparing the change in clinical profiles from the time of PA sensor implant to the time of LVAD implantation, there were important differences between the treatment and control groups. Changes in creatinine, GFR, and blood urea nitrogen (BUN) were minimal in the treatment group, whereas changes in the control group trended toward worsening renal function. There was a trend toward a shorter length of time in the treatment group to proceed to LVAD implantation (168.6 ± 131.8 versus 265.7 ± 210.3 days; HR, 1.72; 95% [confidence interval] CI, 0.80–3.70; p = 0.13) (Figure 2).
During the randomized follow-up and before LVAD implantation, the treatment group tended to have more HF medication changes than did the control group (10.8 ± 6.0 vs. 8.3 ± 8.4; p = 0.14). Although this was not a statistically significant difference, it is important to note that the time period when these medication changes were made was shorter in the treatment group than the control group indicating a higher concentration of therapy changes in the treatment group when physicians have access to hemodynamic information. Interestingly, medication changes made in both the treatment and control groups did not result in lower cardiac filling pressures. These medication data are provided in Table 4. The average INTERMACS profile score for the treatment and control groups was 3. Only two patients in the treatment group (13%) and two patients in the control group (17%) were INTERMACS profile 2.
During the post-LVAD time period, the treatment group experienced large reductions in PA systolic pressure (−16.5 ± 12.4 mm Hg), PA diastolic pressure (−9.4 ± 7.8 mm Hg), and PA mean pressure (−11.6 ± 9.1 mm Hg). An example of a patient’s PA pressure readings leading up to LVAD implantation, after and subsequently when the patient received cardiac transplantation, is shown in Figure 3. The magnitude of each of these reductions was statistically significant (p < 0.01). In contrast, the control group experienced decreases in PA pressures after LVAD implantation as expected, but these reductions were smaller in magnitude when compared with the treatment group and were not statistically significant. Specifically, the control group PA systolic pressure was reduced by −8.9 ± 12.5 mm Hg (p = 0.11), PA diastolic pressure was reduced by −6.3 ± 7.3 mm Hg (p = 0.07), and PA mean pressure was reduced by −7.2 ± 8.5 mm Hg (p = 0.07). During the period after LVAD implantation, the treatment group underwent slightly more HF medication changes than did the control group, but these results were not statistically significant.
Analysis of time from LVAD implantation to heart transplantation or study end in Figure 4 showed a significantly shorter length of time in the treatment group (202.0 ± 162.8 vs. 284.8 ± 153.8 days; HR, 9.53; 95% CI, 2.34 – 38.73; p < 0.01). In addition, the proportion of patients bridging to transplant was significantly greater for the treatment group compared with control (seven treatment patients versus one control patient, p = 0.043). The authors acknowledge that the decision and ability to proceed with transplantation for a given patient is dependent on many complex variables that are beyond the scope of this limited patient experience. Further investigations into this observation in larger patient groups are needed.
Of the 15 treatment patients who underwent LVAD implantation, seven (47%) were successfully bridged to transplantation, two patients died, and the remaining six continued LVAD support at the end of follow-up. Of the 12 control patients who underwent LVAD implantation, one (8%) patient was successfully bridged to transplantation, one patient died, and the remaining 10 patients are continued LVAD therapy at the end of follow-up.
There are several interesting observations from this initial experience in a subset of patients who underwent LVAD implantation in the CHAMPION Trial. In contrast to other intracardiac sensors which may interfere with the positioning or implant of the LVAD, the sensor used in the CHAMPION Trial is implanted in an extracardiac location. Patients were able to upload high-quality pulsatile PA pressure tracings without reported interference from the LVAD.
The CHAMPION Trial provided an opportunity to observe the transition from NYHA Class III symptoms to advanced HF that required LVAD placement. Twenty seven patients (5%), all with NYHA Class III symptoms at the time of enrollment in the study, deteriorated to the point that implantation of an LVAD was necessary. Although these patients were clinically assessed to have NYHA Class III, american college of cardiology/ american heart association (ACC/AHA) Stage C HF at the time of implantation, they had higher PA pressures, lower systemic systolic pressures, and higher creatinine levels. Furthermore, medication changes made during the randomization period were not effective in lowering PA pressures before LVAD implantation. After PA sensor implantation, patients had more medication changes in both the treatment and control groups, but little response in PA pressures was seen. This pattern may provide earlier insight into “medical futility” as patients progress toward the need for advanced therapies. This suggests that hemodynamic characterization, using an implanted device, may provide insight into the natural progression of HF disease to advanced therapy status. More study of this concept is warranted and may be very useful for proper recognition of underlying worsening and transition from Stage C to advanced Stage D HF.
Although the number of clinically indicated LVAD implantations were similar in both the treatment and control groups and appropriate HF drug and device treatments were at optimal or best-tolerated stable doses at baseline, the information obtained from serial monitoring of PA pressures appeared to have a clinical impact in both the pre- and post-LVAD periods. At the time of PA sensor implant, all patients had similar clinical profiles and were receiving similar medical management therapies. Before LVAD implantation, the treatment group underwent more HF medication changes than did the control group in an attempt to alter their hemodynamic status. Access to hemodynamic information in the treatment group during this time period may have allowed physicians to determine patient nonresponsiveness to pharmacologic therapies in a manner unavailable to patients in the control group. Frequent hemodynamic evaluation and the ability to track the effect of different pharmacologic therapies on hemodynamics may enable physicians to reach the conclusion sooner that LVAD therapy is required, and this concept is a plausible explanation for why the treatment group trended toward earlier LVAD implantation.
Proper timing of LVAD implantation continues to be a subject of intense interest because an appropriate measure for the best timing is lacking. Several risk scores and models have also been developed,17,18 all based on retrospective analysis of trials or registries, with the premise that with proper patient selection and earlier device placement, one can reduce the morbidity and mortality associated with LVADs. Data from the INTERMACS have suggested that better results can be obtained in patients who are not as compromised at the time of LVAD implantation.19 A review of INTERMACS profiles revealed that 80% of current devices are being used in two profiles with the highest levels of clinical compromise (1 and 2). In the CHAMPION study LVAD cohort, the average INTERMACS profile score for the treatment and control groups was 3. Only two patients in the treatment group (13%) and two patients in the control group (17%) were INTERMACS profile 2. This is an important consideration and shows the high-quality standard of care in the control group provided by the centers participating in the study. Despite this fact, the treatment group still underwent earlier LVAD implantation and experienced more effective pressure and volume unloading of the LV compared with patients in the control group. Physician knowledge of hemodynamic information may have allowed for more dynamic and patient-specific management of the LVAD device settings leading to better LVAD device performance. This observation was anecdotally described by physicians in the CHAMPION trial who had direct experience managing both treatment and control patients undergoing LVAD implantation. Further studies are needed to evaluate how hemodynamic monitoring may impact outcomes in patients who receive VAD therapy.
There are important limitations to this study. The analyses described are all post hoc and included a relatively small number of patients in each group. The results observed may not be representative of the larger patient population who are in need of LVAD implantation. However, these findings and the potential impact of PA pressure monitoring in LVAD patients are encouraging. This new approach could be superior to other device-based measurements (e.g., pulsatility index, estimated LVAD flows, and periodic echocardiography) in helping physicians make medication adjustments involving neurohormonal antagonists, diuretics, and vasodilators as well as in optimizing LVAD device performance.
Finally, in this study, we found that more patients in the treatment group were transplanted and at an earlier time point. Unquestionably, transplantation is dependent on many complex variables including organ availability, significant geographic variation in organ transplantation, and other issues related to donor compatibility. Additional studies in larger LVAD patient populations are needed to determine what effect implantable hemodynamic monitoring systems may have on transplant status.
More clinical trials are required to address how PA pressure monitoring may impact outcomes in patients who receive advanced therapies. Two possible future investigations would include a study to determine the effectiveness of prospectively identifying the ideal timing for LVAD implantation, while a second would prospectively focus on the management of intracardiac filling pressures while on LVADs and the impact on LVAD device performance. This second study will become key as there are increases in outpatient LVADs, greater distances from implanting centers, and longer periods of survival after LVAD placement.
The findings reported in this retrospective analysis are important because they suggest that monitoring PA pressures may identify patients in need of LVAD at an earlier time point. In addition, this technology may lead to the possible improvement of patient care while being maintained on a LVAD.
1. Stevenson LW, Zile M, Bennett TD, et alChronic ambulatory intracardiac pressures and future heart failure
events. Circ Heart Fail 2010.3: 580–587,
2. Abraham WT, Adamson PB, Bourge RC, et alCHAMPION Trial Study Group: Wireless pulmonary artery haemodynamic monitoring in chronic heart failure
: A randomised controlled trial. Lancet 2011.377: 658–666,
3. Bourge RC, Abraham WT, Adamson PB, et alCOMPASS-HF Study Group: Randomized controlled trial of an implantable continuous hemodynamic monitor in patients with advanced heart failure
: The COMPASS-HF study. J Am Coll Cardiol 2008.51: 1073–1079,
4. Abraham WT, Stevenson LW, Bourge RC, Lindenfeld JA, Bauman JG, Adamson PBCHAMPION Trial Study Group: Sustained efficacy of pulmonary artery pressure to guide adjustment of chronic heart failure
therapy: Complete follow-up results from the CHAMPION randomised trial. Lancet 2016.387: 453–461,
5. Costanzo MR, Stevenson LW, Adamson PB, et alInterventions linked to decreased heart failure
hospitalizations during ambulatory pulmonary artery pressure monitoring. JACC Heart Fail 2016.4: 333–344,
6. Adamson PB, Abraham WT, Bourge RC, et alWireless pulmonary artery pressure monitoring guides management to reduce decompensation in heart failure
with preserved ejection fraction. Circ Heart Fail 2014.7: 935–944,
7. Adamson PB, Abraham WT, Stevenson LW, et alPulmonary artery pressure-guided heart failure
management reduces 30-day readmissions. Circ Heart Fail 2016.9:e002600,
8. Adamson PB, Abraham WT, Aaron M, et alCHAMPION trial rationale and design: The long-term safety and clinical efficacy of a wireless pulmonary artery pressure monitoring system. J Card Fail 2011.17: 3–10,
9. Adamson PB, Ginn G, Anker SD, Bourge RC, Abraham WTRemote haemodynamic-guided care for patients with chronic heart failure
: A meta-analysis of completed trials. Eur J Heart Fail 2017.19: 426–433,
10. Ponikowski P, Voors AA, Anker SD, et alAuthors/Task Force Members: 2016 ESC guidelines for the diagnosis and treatment of acute and chronic heart failure
: The task force for the diagnosis and treatment of acute and chronic heart failure
of the European Society of Cardiology (ESC). Developed with the special contribution of the Heart Failure
Association (HFA) of the ESC. Eur Heart J 2016.37: 2129–2200,
11. Miller LW, Pagani FD, Russell SD, et alHeartMate II Clinical Investigators: Use of a continuous-flow device in patients awaiting heart transplantation
. N Engl J Med 2007.357: 885–896,
12. Slaughter MS, Rogers JG, Milano CA, et alHeartMate II Investigators: Advanced heart failure
treated with continuous-flow left ventricular assist device
. N Engl J Med 2009.361: 2241–2251,
13. Hoppe UC, Vanderheyden M, Sievert H, et alChronic monitoring of pulmonary artery pressure in patients with severe heart failure
: Multicentre experience of the monitoring Pulmonary Artery Pressure by Implantable device Responding to Ultrasonic Signal (PAPIRUS) II study. Heart 2009.95: 1091–1097,
14. Hunt SA, Abraham WT, Chin MH, et alAmerican College of Cardiology Foundation; American Heart Association: 2009 focused update incorporated into the ACC/AHA 2005 guidelines for the diagnosis and management of heart failure
in adults a report of the american college of cardiology foundation/american heart association task force on practice guidelines developed in collaboration with the international society for heart and lung transplantation. J Am Coll Cardiol 2009.53: e1–e90,
15. Stevenson LW, Pagani FD, Young JB, et alINTERMACS profiles of advanced heart failure
: The current picture. J Heart Lung Transplant 2009.28: 535–541,
16. Matthews JC, Koelling TM, Pagani FD, Aaronson KDThe right ventricular failure risk score a pre-operative tool for assessing the risk of right ventricular failure in left ventricular assist device
candidates. J Am Coll Cardiol 2008.51: 2163–2172,
17. Ketchum ES, Moorman AJ, Fishbein DP, et alPredictive value of the Seattle heart failure
model in patients undergoing left ventricular assist device
placement. J Heart Lung Transplant 2010.29: 1021–1025,
18. Kirklin JK, Naftel DC, Kormos RL, et alSecond INTERMACS annual report: More than 1,000 primary left ventricular assist device
implants. J Heart Lung Transplant 2010.29: 1–10,
19. Kirklin JK, Naftel DC, Kormos RL, et alThird INTERMACS Annual Report: The evolution of destination therapy in the United States. J Heart Lung Transplant 2011.30: 115–123,
Keywords:Copyright © 2018 by the American Society for Artificial Internal Organs
hemodynamic monitoring; mechanical circulatory support; left ventricular assist device; heart failure; pulmonary wedge pressure; heart transplantation