Continuous-flow left ventricular assist devices (CF-LVADs) used in advanced heart failure (HF) are set at a fixed pump speed, which is based, primarily, on echocardiographic documentation of adequate, but not excessive, left ventricular (LV) unloading at the particular speed.1,2 Ramp studies, measuring changes in cardiac parameters as a function of serial pump speed changes, are increasingly being used as an indicator of pump function and malfunction in CF-LVAD patients. A low degree of reduction in LV diameter during pump speed increase may indicate pump dysfunction, inflow cannula malposition, excessive unloading, hypovolemia, or low LV elasticity. In contrast, a larger change in LV dimensions during pump speed alterations has been suggested to indicate optimal function of the pump and of the device–heart interaction.3–6 The ability to predict functional status and outcome from ramp studies remains unexplored. The aim of this study was to evaluate the clinical relevance of ramp tests by analyzing invasive hemodynamic changes during ramp testing and by assessing the relation of such changes to functional capacity, quality of life (QOL), and outcome in CF-LVAD patients.
Materials and Methods
Patient Cohort and Design
In this single-center study, 44 CF-LVAD patients, all HeartMate II (Thoratec Corporation, Pleasanton, CA), were studied. Between 2006 and 2009, the patients underwent 80 ramp tests (1.8 ± 1.0 per patient) measuring hemodynamic changes by right heart catheterization (RHC) as a function of pump speed (revolutions per minute [rpm]). The ramp studies were performed as part of the routine clinical evaluation of hemodynamics post implant and all eligible patients in the inclusion period were tested and included in this study. Approximately 2 weeks after the ramp study, functional capacity, health status, and QOL were assessed. In patients with more than one ramp test performed, the test with the closest proximity to functional evaluation was chosen. Functional capacity was assessed by 6 minute walk test (6MWT), New York Heart Association (NYHA) functional classification, and activity score (five categories; very low = 1, low = 2, moderate = 3, high =4, very high = 5). Health status and QOL were assessed through two questionnaires; Kansas City Cardiomyopathy questionnaire (KCCQ) and Minnesota Living with Heart Failure (MLWHF). Kansas City Cardiomyopathy status was described both by the overall summary score, combing scores in all domains (KCCQa) and by the clinical summary score, combing physical and symptoms scores only (KCCQb).
Ramp Speed Protocol
Hemodynamic measurements were undertaken at three different pump speeds in all patients: 1) at the baseline pump speed the patient came in with, 2) at maximal pump speed, and 3) at low pump speed. Some patients had additional measurements done between these three ramp speed steps. The ramp protocol was initiated at the patient’s usual pump speed (~ramp-base). Subsequently measurements were done at the highest possible pump speed (~ramp-high), which was decided bedside by the physician based on LV dimensions and/or occurrence of septal shift or arrhythmias. Finally, pump speed was decreased, if possible, to the point of aortic valve (AV) opening or to 8,000 rpm (~ramp-low). Before taking any measurements at the next pump speed, a minimum of 3 min was allowed to reach equilibrium.
At each pump speed central pressures, cardiac output (CO) and pulmonary artery saturation (PA sat) were measured, on full anticoagulant therapy, by RHC via the right internal jugular vein, using a Swan-Ganz catheter (PAC VIP, Edwards Lifesciences, Irvine, CA). Cardiac output was measured three times at each ramp step by standard thermodilution using cold saline injection, and the average result was recorded.
Echocardiography was performed according to current guidelines4,7 on a General Electric ultrasound system. Pulsed and continuous wave Doppler images were acquired in the parasternal long-axis view and the apical four chamber view. Left ventricular dimensions were measured from parasternal images. Aortic valve opening was assessed in both the parasternal long-axis view and the apical view. At least 10 consecutive cardiac cycles were reviewed, and the frequency of AV opening recorded as a proportion.
At each ramp step pump speed, pulsatility index, pump flow and power consumption were noted.
Data were analyzed using SAS 9.4 Statistical Software (Cary, NC). Because patients had varying ranges of pump speed change from ramp-low to ramp-high, data were presented as hemodynamic change per rpm (~Δchange/Δrpm)—thus allowing for comparisons of between-patient differences in response to pump speed change. Relations between continuous data (KCCQ, MLWHF, and 6MWT) and hemodynamic variables were described by correlation coefficients. One-way ANOVA was used to analyze the relation between categorical data (activity score and NYHA functional classification) and hemodynamic variables.
Chi-square tests were used to analyze the relationship between categorical variables; the Fisher’s exact test was applied when one or more of the cells had an expected frequency of five or less. Differences between groups were examined using two-sided Student’s t test, paired, or unpaired as appropriate. Mortality rates were evaluated using Kaplan–Meier curves and differences between groups were tested using the log-rank method. When analyzing survival, patients who were heart transplanted (HTX) were censored at that time. Statistical significance was defined by a two-tailed p value < 0.05. Data are presented as mean ± standard deviation unless otherwise stated.
Patient characteristics are shown in Table 1. The cohort predominantly consisted of male patients suffering from nonischemic cardiomyopathy. In 80% of the cases, the CF-LVAD was utilized as bridge to transplantation (BTT). At the day of ramp testing, support duration was 73 ± 72 days. At the average baseline pump speed (9,605 ± 362 rpm), the AV was closed in 35, open in 7, and intermittently open in 2 patients. As expected pump speed in these three AV groups differed significantly; 9,680 ± 357, 9,286 ± 227, and 9,400 ± 0 rpm, respectively (p = 0.02).
Hemodynamic Changes as a Function of Pump Speed
During the ramp studies, pump speed was increased 23% (1,964 ± 584 rpm) from the lowest to the highest pump speed (ramp-low 8,609 ± 518 rpm vs. ramp-high 10,573 ± 555 rpm) without complications (Table 2). Cardiac output increased 1.1 ± 1.0 l/min from lowest to highest pump speed corresponding to 0.6 ± 0.6 ml/min/rpm; ramp-low 4.9 ± 1.2 vs. ramp-high 6.0 ± 1.4 L/min (p < 0.001). Pulmonary capillary wedge pressure (PCWP) decreased 5.8 ± 6.1 mm Hg from lowest to highest pump speed corresponding to −0.28 ± 0.19 mm Hg/rpm * 10−2; ramp-low 14.5 ± 7.3 vs. ramp-high 8.7 ± 6.1 mm Hg (p < 0.001). Pulmonary artery saturation increased 6.1 ± 4.0% from lowest to highest pump speed corresponding to 0.29 ± 0.2%/rpm * 10−2; ramp-low 63.3 ± 6.5 vs. ramp-high 69.4 ± 65.4% (p < 0.001).
Following ramp testing, 18 patients had their pump speed changed for optimization, nine had pump speed increased (378 ± 120 rpm), and nine had pump speed decreased (267 ± 100 rpm).
Functional Capacity and QOL
Functional capacity and QOL in relation to ramp testing was estimated in 31/44 patients (70%) 15 ± 12 days post ramp. Average 6MWT was 312 ± 220 min, activity score was 2.5 ± 0.9 and NYHA functional class was I–II in 70% and III–IV in 30% (one patient in NYHA class IV). Quality of life scores assessed with MLWHF reached 52 ± 27 U, 50 ± 21 U in KCCQa, and 59 ± 21 U in KCCQb. Missing data (n = 13) were due to patient/staff unavailable to assess parameters, >60 days between ramp test and evaluation, death or HTX.
Relation Between Ramp Tests and Functional Assessment
Changes in CO and PCWP as a function of delta pump speed and their relation to functional capacity and QOL are shown in Table 3. Decrease in PCWP as a function of pump speed (ΔPCWP/Δrpm) was associated with higher activity scores; very low–low vs. moderate–very high −0.16 ± 0.16 vs. −0.31 ± 0.16 mm Hg/rpm * 10–2 (p = 0.02), respectively. Patients in NYHA functional class I–II vs. III–IV had a significantly larger decline in PCWP/Δrpm; −0.29 ± 0.15 vs. −0.09 ± 0.16 mm Hg/rpm * 10–2 (p = 0.007), respectively, and a larger increase in CO/Δrpm; 0.65 ± 0.63 vs. 0.19 ± 0.31 (p = 0.01) respectively (Figure 1). Furthermore, increase in CO/Δrpm was significantly associated with favorable QOL; MLWHF score (R = −0.42, p =0.03), KCCQa (R = 0.46, p = 0.01), and KCCQb (R = 0.46, p = 0.01).
Baseline Pump Speed and Functional Assessment
At baseline pump speed, cardiac index (CI) correlated moderately with 6MWT, R = 0.39 (p = 0.04). Patients with activity scores “very low–low” (55%) had a lower CI than “moderate–very high” patients (45%), 2.8 ± 0.5 vs. 3.0 ± 0.6 L/min/m2, however nonsignificant. Furthermore, patients in NYHA class I–II tended to have a higher CI than NYHA III–IV patients; 3.0 ± 0.6 vs. 2.8 ± 0.5 L/min/m2, but did not reach statistical significance either. Finally, CI at baseline pump speed was correlated with measures of QOL; MLWHF, KCCQa, and KCCQb, R = −0.42 (p = 0.037), R = 0.44 (p = 0.02), and R = 0.45 (p = 0.02), respectively.
Left Ventricular End-Diastolic Diameter
Cardiac output increase/Δrpm in patients with baseline left ventricular end-diastolic diameter (LVEDD) ≥ median (5.1 cm) was 0.67 ± 0.59 ml/min/rpm vs. 0.61 ± 0.42 ml/min/rpm in patients with baseline LVEDD < median; p = 0.8. Pulmonary capillary wedge pressure decline/Δrpm in patients with baseline LVEDD ≥ median was −0.42 ± 0.18 mm Hg/rpm * 10–2 vs. −0.29 ± 0.22 mm Hg/rpm * 10–2 in patients with baseline LVEDD < median; p = 0.12. Hence, the hemodynamic response to changes in pump speed did not depend on baseline LV diameter.
Right Ventricular Function and Changes in ΔCO/rpm
Increase in CO/ΔRPM in patients below versus above median right atrial pressure did not differ significantly; 0.33 ± 0.28 vs. 0.66 ± 0.73 ml/min/rpm (p = 0.1) nor between patients with RVEDD below versus above median; 0.67 ± 0.75 vs. 0.39 ± 0.4 ml/min/rpm (p = 0.2).
Age and Functional Capacity
Patients younger than median (50 years) tended to walk longer in 6MWT than older patients; 389 ± 238 vs. 242 ± 181 min (p = 0.08). Forty-seven percent of patients younger than median had a “very low–low” activity score, whereas this was the case in 64% of the older (p = 0.46). Also NYHA functional classification tended to be preferable in the younger patients; 80% of patients below the median age were in NYHA class I–II versus only 60% of patients above the median age (p = 0.43). There was no difference in changes in CO/Δrpm and PCWP/Δrpm between younger and older patients.
Median follow-up in survivors was 2,251 days (range: 1,781–2,770) since the ramp study. In total, 32% (n = 14) had undergone HTX after waiting 586 ± 335 days. The composite of transplanted survivors (n = 8) and ongoing destination therapy (DT) support patients (n = 7) were 34%, while 66% (n = 29) died. Out of the 29 dead, two were DT patients, six had previously been transplanted and 21 died on the waiting list. Survival in patients with decrease in PCWP/Δrpm above versus below median during ramp testing was not significantly different (p = 0.66), Figure 2. Nor was survival between patients with CO increase/Δrpm above versus below median (p = 0.99).
Data from the clinical trials have documented that CF-LVAD implantation enhances exercise tolerance and QOL in HF patients.8,9 The relation between functional capacity and central hemodynamic status, however, has not been well described in HF patients treated with mechanical circulatory support.
This study shows that baseline hemodynamic parameters in general are poor predictors of measures of QOL and functional capacity in CF-LVAD patients. Only baseline CI was moderately correlated with 6MWT and QOL. Consistently in unsupported HF patients, hemodynamic variables at rest generally are poor indicators of functional capacity unless cardiac function is severely reduced. In contrast the correlation, in HF patients during exercise, between CO and functional capacity is highly significant.10,11 Thus, the lack of correlation between hemodynamic variables and functional status at rest demonstrated in the current study is consistent with the findings in HF patients in general and supports the hypothesis that several cardiovascular responses typical for HF patients (HF phenotype) remain in LVAD-treated patients.
Ramp studies, testing changes in cardiac variables, such as dimensions or pressure, as a function of pump speed, are commonly used to optimize pump speed settings and to detect pump obstructions or dysfunction.3–6 The ability in CF-LVAD patients to respond hemodynamically to increasing pump speed might be compared with cardiac reserve in healthy individuals. The analysis showed that CO increase in response to increased pump speed was related to improved QOL and better NYHA classification. The explanation for this finding is not completely clear. We did not find right atrial pressure below or above median to be associated with degree of CO increase/Δrpm. However, to evaluate the impact of RV function on exercise capacity in CF-LVAD recipients, a dedicated prospective study is warranted using current guidelines12 including more than one measure, such as RV fractional area change (RVFAC), tricuspid annular plane systolic excursion (TAPSE), and pulsed tissue Dopplers’. In the absence of shunts, total CO will equal RV output. Thus, it may be that ability of the right ventricle to accommodate and pump forward the increased preload generated by the initial increase in pump outflow in reality reflects better RV function during hemodynamic stress in the patients who responded positively in terms of CO in the ramp test. If so, this would be consistent with previous data from conventionally treated HF patients showing that RV function is a predictor of functional capacity.13,14 However, further studies are required to fully determine the mechanism behind the correlation.
Greater degrees of unloading of the LV (~decreasing PCWP) with increasing pump speed was found to be associated with lower NYHA functional classification and higher activity scores. In concordance, filling pressures in unsupported HF patients seem to be correlated with functional capacity.15 Possibly, the decrease in PCWP during pump speed increase reflects that the LVAD cannula is well positioned and that, consequently, the pump is able to unload effectively the LV, and prevent dyspnea, during increased hemodynamic demands. This would explain the better exercise tolerance in patients where PCWP responded favorably to increases in pump speed during ramp tests.
As expected, although not statistically significant, functional capacity showed a tendency toward declining with older age. Activity scores, 6MWT, and symptoms exceeding the median NYHA functional classification of the cohort were disadvantageous in older compared with younger CF-LVAD patients. This relation between age and exercise capacity is well known in both mechanically supported and unsupported HF patients and can be explained by multiple parameters, such as declining pulmonary capacity and muscle wasting in the elderly.1 We found no differences in hemodynamic responses to pump speed increase when comparing younger and older patients.
In contrast to our hypothesis, the results of the ramp tests in terms of hemodynamic measurements did not predict survival. Lack of increase in CO or lack of reduction in PCWP during ramp testing, in theory, indicates suboptimal function of the LVAD, which should be expected to lead to poorer outcome. A number of factors may explain the lack of such an association in the current study. Possibly the test variable above versus below the median for change in PCWP and CO is too insensitive to detect LVADs with suboptimal function. However, given the limited number of patients, it is impossible in the current study with any confidence to analyze the outcome of more extreme groups. Second, patients with pump dysfunction or persistent severe HF symptoms may have been offered urgent or semiurgent transplantation, which would likely add bias to the comparison of groups. Preferably, this hypothesis should be tested in a larger cohort restricted to patients treated with the device as DT.
A number of limitations associated with this study should be mentioned. First, the study was retrospective and the sample size small. Second, the heterogeneity of the cohort including differences in HF etiology, age, LVAD indication, and gender requires caution when interpreting findings. Nevertheless, this study was too small to carry out multifactorial statistical analysis. The relatively short support duration might be insufficient to define the cohort as overall stable after VAD implantation. Furthermore, ramp tests and status evaluation were not performed on the same day, but were, on average, separated by approximately 2 weeks. Thus, an abrupt change in health conditions could potentially have affected the association between hemodynamic and physical parameters in some patients. However, the likelihood of rapid changes in the patient’s conditions within such short time is small and if detrimental, patients would not have been able to complete tests evaluating their functional status. Due to the described limitations, findings must be interpreted with caution and larger prospective studies are warranted to test the hypothesis of the study.
Hemodynamic responses during ramp testing in CF-LVAD recipients are associated with measures of functional capacity and QOL. We found that patients in NYHA class I–II had a significantly larger increase in CO/ΔRPM and decrease in PCWP/ΔRPM compared with NYHA III–IV patients. Furthermore, increase in CO/ΔRPM was associated with QOL. Hence, ramp studies may identify patients likely to thrive during CF-LVAD support with an acceptable QOL and exercise tolerance. Hemodynamic responses during ramp testing were not correlated to survival in this study; however, larger studies are required to test this hypothesis.
1. Jung MH, Gustafsson F: Exercise in heart failure patients supported with a left ventricular assist device. J Heart Lung Transplant 2015.34: 489–496.
2. Jung MH, Hansen PB, Sander K, et al: Effect of increasing pump speed during exercise on peak oxygen uptake in heart failure patients supported with a continuous-flow left ventricular assist device. A double-blind randomized study. Eur J Heart Fail 2014.16: 403–408.
3. Uriel N, Morrison KA, Garan AR, et al: Development of a novel echocardiography ramp test for speed optimization and diagnosis of device thrombosis in continuous-flow left ventricular assist devices: The Columbia ramp study. J Am Coll Cardiol 2012.60: 1764–1775.
4. Estep JD, Stainback RF, Little SH, Torre G, Zoghbi WA: The role of echocardiography and other imaging modalities in patients with left ventricular assist devices. JACC Cardiovasc Imaging 2010.3: 1049–1064.
5. Kato TS, Colombo PC, Nahumi N, et al: Value of serial echo-guided ramp studies in a patient with suspicion of device thrombosis after left ventricular assist device implantation. Echocardiography 2014.31: E5–E9.
6. Jung MH, Hassager C, Balling L, Russell SD, Boesgaard S, Gustafsson F: Relation between pressure and volume unloading during ramp testing in patients supported with a continuous-flow left ventricular assist device. ASAIO J 2015.61: 307–312.
7. Cheitlin MD, Armstrong WF, Aurigemma GP, et al; ACC; AHA; ASE: ACC/AHA/ASE 2003 Guideline Update for the Clinical Application of Echocardiography: Summary article. A report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (ACC/AHA/ASE Committee to Update the 1997 Guidelines for the Clinical Application of Echocardiography). J Am Soc Echocardiogr 2003.16: 1091–1110.
8. Rogers JG, Aaronson KD, Boyle AJ, et al; HeartMate II Investigators: Continuous flow left ventricular assist device improves functional capacity and quality of life of advanced heart failure patients. J Am Coll Cardiol 2010.55: 1826–1834.
9. Kirklin JK, Naftel DC, Pagani FD, et al: Seventh INTERMACS annual report: 15,000 patients and counting. J Heart Lung Transplant 2015.34: 1495–1504.
10. Cooke GA, Marshall P, al-Timman JK, et al: Physiological cardiac reserve: Development of a non-invasive method and first estimates in man. Heart 1998.79: 289–294.
11. Sullivan MJ, Knight JD, Higginbotham MB, Cobb FR: Relation between central and peripheral hemodynamics during exercise in patients with chronic heart failure. Muscle blood flow is reduced with maintenance of arterial perfusion pressure. Circulation 1989.80: 769–781.
12. Rudski LG, Lai WW, Afilalo J, et al: Guidelines for the echocardiographic assessment of the right heart in adults: A report from the American Society of Echocardiography endorsed by the European Association of Echocardiography, a registered branch of the European Society of Cardiology, and the Canadian Society of Echocardiography. J Am Soc Echocardiogr 2010.23: 685–713; quiz 786.
13. Di Salvo TG, Mathier M, Semigran MJ, Dec GW: Preserved right ventricular ejection fraction predicts exercise capacity and survival in advanced heart failure. J Am Coll Cardiol 1995.25: 1143–1153.
14. Murninkas D, Alba AC, Delgado D: Right ventricular function and prognosis in stable heart failure patients. J Card Fail 2014.20: 343–349.
15. Smart N, Haluska B, Leano R, Case C, Mottram PM, Marwick TH: Determinants of functional capacity in patients with chronic heart failure: Role of filling pressure and systolic and diastolic function. Am Heart J 2005.149: 152–158.
Keywords:Copyright © 2016 by the American Society for Artificial Internal Organs
left ventricular assist device; hemodynamics; functional capacity; quality of life; heart failure