Acharya, Deepak*; Singh, Satinder†; Tallaj, José A.*; Holman, William L.‡; George, James F.‡; Kirklin, James K.‡; Pamboukian, Salpy V.*
Heart failure is an increasingly common problem.1 An aging population coupled with advances in cardiovascular care has resulted in a growing number of older patients with advanced heart failure. Heart transplantation, although being highly effective, remains limited by donor organ shortages, with 2,163 patients receiving heart transplants in the United States in 2008.2
Left ventricular assist devices (LVADs) have assumed an increasingly important role in the management of advanced heart failure, especially after the landmark REMATCH trial that demonstrated meaningful benefit in both survival and quality of life.3 Since then, improved patient selection and advances in technology have led to further improved outcomes with reported 1-year survival of 77%–82%.4–6 As the horizon of support with LVADs increases, multiple complications may develop, including persistent heart failure symptoms, device malfunction, thrombus formation, bleeding, hemolysis, neurologic events, and sepsis.
Traditionally, transthoracic (TTE) and transesophageal echocardiography (TEE) have been used to provide important structural and hemodynamic information in patients with LVADs. Transthoracic echocardiography, although noninvasive, is limited by body habitus, postsurgical changes, inability to visualize apex, respiratory changes, and limited image quality in some patients. Structural information, such as device cannula position, may not be directly observed and is inferred from the patterns of blood flow. With careful technique, the yield may be higher,7 but significant interoperator variability and technical problems remain. Transesophageal echocardiography, although more invasive, is very useful in the assessment of LVAD and the heart. Left ventricular assist device position, adequacy of deairing during LVAD insertion, shunts, left ventricular (LV) and right ventricular (RV) function, valvular structure and hemodynamics, thrombus, and aortic root can be studied with TEE.8 Catheter-based approaches to assessment of LVAD function have been described, including right heart catheterization, LVAD cannula catheterization in pulsatile pumps, and angiography,9,10 but these are not standardized, may be not possible with continuous flow devices and are invasive.
In this report, we describe our experience using gated CTA to assess LVAD function in selected patients with suspected LVAD dysfunction.
Fourteen patients with a LVAD followed at the University of Alabama Hospital (UAB) underwent contrast-enhanced gated computed tomography (CT) angiogram of the chest between September 2007 and May 2009. Indications for CT angiogram included persistent heart failure symptoms, evidence of RV dysfunction, discrepancy between LVAD flows and cardiac output, recurrent LVAD alarms, hemolysis, and symptomatic hemodynamic instability. Parameters collected were the type and duration of cardiomyopathy, LVAD type, onset of clinical problem, alternative studies to evaluate problem, results of CTA, change in management, and outcomes. The range and median duration that the LVADs were in situ when the CTA was performed are 1–28 months and 6.5 months, respectively. During the study period, 45 patients had their first LVAD placed at UAB hospital, and 12 had LVAD exchange.
Technique for CTA
Retrospectively gated contrast-enhanced cardiac CTA was performed using a 64-detector Phillips scanner with the following parameters: collimation, 64 × 0.625 mm, rotation time, 400 ms, pitch, 0.2, tube voltage, 120 kV, tube current, 600–900 mA (depending on patient's body habitus), and slice thickness was 1.4 mm at 0.7-mm increment. Bolus tracking method was used to trigger the scan after reaching a predetermined threshold after intravenous contrast injection. The region of interest (ROI) was placed in the proximal descending thoracic aorta away from the cannulae insertion site to avoid any misregistration. Approximately 80–100 ml of omnipaque 350 contrast followed by 40 ml of saline chaser was administered at 4.5 ml/s injection rate, and scans were acquired in craniocaudal direction from thoracic inlet to below the LVAD cannulae. Electrocardiographic (ECG) dose modulation was used during systolic phase of the cardiac cycle to reduce radiation exposure. Patients were not given beta-blockers or nitroglycerine. All 10 phases of CTA dataset was reconstructed at 10% interval (0%–90%), and images were further analyzed using Brilliance 2.1 Philips as well as TeraRecon workstations. All cardiac CTA studies were interpreted by an experienced radiologist. The inflow cannula position was defined as normal if it was directed toward the mitral valve and not abutting the LV wall.
Fourteen patients with ventricular assist devices had gated CTA to evaluate various clinical concerns. Age at LVAD placement ranged from 18 to 71 years, with a median age of 56 years. There were 11 men and 3 women. Five of the 14 patients had ischemic cardiomyopathy. Of the nine who had nonischemic cardiomyopathy, one had postpartum cardiomyopathy, one had adriamycin-induced cardiomyopathy, and the cause of cardiomyopathy in the remaining seven was unknown. Initial devices implanted were HeartMate II LVAD in six, HeartMate XVE in three, VentrAssist in four, and AbioMed AB5000 in one.
Clinical concerns included persistent heart failure symptoms in six patients, hemodynamic issues manifested as low flow alarms or dizziness in six patients, hemolysis in one patient, and abnormal LVAD sounds in one patient. Creatinine values on the date of the CTA ranged from 0.7 to 1.6 mg/dl (median 0.8 mg/dl).
For 13 of 14 patients, transthoracic echocardiograms were available for review. Poor biventricular function was observed in nine patients, poor LV function in two patients, and normalized LV function in one patient. One study was very limited due to poor windows. Right heart catheterization results were available in six patients. Normal hemodynamics were measured in one patient, elevated filling pressures with normal cardiac index were measured in three patients, and elevated filling pressures with depressed cardiac index were observed in two patients.
Computed tomography angiograms did not demonstrate any abnormality of the LVAD or associated structures in 6 of 14 patients. Normal cannula position is illustrated in Figure 1. With adjustments in medical management or monitoring only, these patients remained stable with no serious sequelae. One patient died due to non-VAD–related issues 2 years later.
Among the eight patients with abnormal CTA, six patients were found to have a malpositioned LV inflow cannula (Figures 2 and 3). In four patients, this was significant enough to warrant exchange of the device. Three patients did well, and one patient died of multiorgan failure after device exchange. This was thought to be due to irreversible right heart failure as a sequelae of suboptimal LV unloading with chronically elevated left-sided pressures. One patient with cannula malposition transitioned to heart transplantation. The sixth patient has done well with adjustment of medications likely because the cannula malposition was minor and not significantly restricting blood inflow to the device.
In a patient with neointimal hyperplasia in the outflow cannula no device exchange was necessary. In one patient, a significant outflow graft stenosis near the aortic anastamosis was identified, which had manifested as dyspnea and severe hemolysis requiring transfusion of multiple units of packed red cells. This narrowing resulted from the attachment of a remnant portion of an AbioMed AB5000 outflow cannula to a newly implanted HeartMate XVE. After the diagnosis was established by CTA, the graft was revised with complete resolution of hemolysis, normalization of hemodynamics, and relief of persistent heart failure symptoms (Figure 4). The clinical characteristics, results of CTA, results of other diagnostics, clinical management, and outcomes of the cohort are outlined in Table 1.
Our study demonstrates that CTA may be safely and effectively used to identify abnormalities in patients with left ventricular assist devices when other diagnostic tools have not been revealing. In our series, a significant proportion of patients had changes in management-based CTA findings.
Persistent or recurrent heart failure symptoms in patients with LVAD have multiple causes. Patient-related and LVAD-related factors should be considered. Evaluation of these patients has traditionally been done with echocardiographic or catheter-based methods. Our report describes the use of CT arteriography of the chest as a promising noninvasive method, which revealed abnormalities not otherwise evident using other diagnostic modalities and led to enhanced patient management.
Transesophageal echocardiography can detect many abnormalities that may affect outcome after LVAD insertion, including dysfunction of the native heart including the valves and right ventricle, inflow or outflow cannula malposition, intracardiac thrombus, LV loading issues, hematoma, and tamponade.8,11–19 Despite the significant utility of TEE, there are limitations. Reverberation signals are common, and there may be linear artifacts or blind spots that cause difficulty in imaging the upper ascending aorta. Also, adequate quantification of aortic regurgitation as well as visualization of the LV apex in a nonforeshortened manner, both crucial in the assessment of LVAD function, can be difficult.20 Anticoagulant management and sedation for TEE are also issues to be considered. Limited evidence shows that TTE done under protocol with special attention to LVAD components can provide valuable information.7 In an observational study of 32 people with HeartMate VE or XVE LVADs, TTE was able identify common problems, such as inflow valve regurgitation, inflow conduit obstruction, outflow graft distortion, and native aortic valve disease and opening. Cardiac catheterization can be used to define causes of persistent heart failure symptoms related to the native heart and VAD-associated issues.21,22 Limitations include its invasive procedure, anticoagulation management periprocedure, and dye-induced nephropathy.
Multidetector computed tomography (MDCT) technology has rapidly evolved over the last few years. With improved spatial and temporal resolution, the clinical applications of gated CTA have expanded beyond the assessment of coronary arteries. In addition to anatomic information, a retrospective gated CTA can provide information regarding left and right ventricular function. Experience using CTA in patients with VADS is limited. In 2000, Smekal et al.23 reported the results of multisclice spiral CT with contrast in a patient 1 week after placement of a DeBakey VAD. Inlet cannula position, outflow graft position and angulation, and the absence of thrombi could be easily detected. As experience has increased, inflow and outflow cannula position and angulation, position of blood pump, driveline position, pericardial effusion, hematoma, and abscess have been diagnosed with CT scan.24 Dynamic CTA has been used to determine cardiac output in patients with continuous-flow LVAD with good correlation between cardiac output by Swan-Ganz thermodilution and cardiac output measurements by CTA at a location in the ascending aorta distal to the anastomosis of the output cannula.25 More recently, a published series demonstrated good sensitivity and specificity of cardiovascular CT compared with intraoperative findings, with poor correlation to LVAD echocardiography.26 Limitations of CTA include need for contrast, radiation exposure, inability to get specific hemodynamic data such as that obtainable by catheterization, and imaging artifacts.
As technology evolves, the role for CTA may expand. Current technology with retrospective gating allows for assessment of LV and RV function, although the cannula can result in artifacts. At this time, thrombus within the pump cannot be confidently evaluated because the pump itself is very dense. However, thrombi in the cannulae can be evaluated.
Our series of 14 patients shows that gated CTA of the chest is feasible and useful in detecting abnormalities in patients with both pulsatile and continuous-flow VADs, which may not be evident using other diagnostic techniques. Issues identified included LVAD cannula malposition and anastamosis. In six of the 14 patients studied, there were direct changes in management as a result of the CTA. In those patients with normal CTA findings relating to the VAD, clinical course was stable, demonstrating a good negative predictive value of a normal CTA. Patients with minor abnormalities did well with clinical monitoring and ongoing medical management.
Further studies to clarify the specific roles of different modalities could include the use of both CTA and TEE when there is relevant clinical concern, and comparison of these results with the surgeon's findings in those patients who are taken to the OR, or with detailed visual examination of the LVAD components in those patients who do not survive or undergo cardiac transplantation. Given the contrast exposure and the underlying renal insufficiency in many end-stage heart failure patients, it may not be possible to do routine CTA in every patient post-LVAD to compare with the intraoperative TEE. A multi-institutional prospective study in suitable patients with specific imaging protocols may allow development of specific parameters and ranges of acceptable normals, which may then be extrapolated to the general LVAD population. Assessment of the LVAD cannula early postoperatively would be useful, but given the clinical instability, renal insufficiency, acquisition times, and transport and logistical issues in many of these patients, early CTA does not have a significant role currently, although this may change with changes in LVAD and CT technology.
There were several limitations to this study. The cohort studied was small, and data were collected retrospectively. Diagnostic studies used for comparison including TEE, TTE, and heart catheterization were done as routine studies by multiple individuals without control for interobserver variability.
As VAD therapy continues to proliferate, challenges will arise in the long-term management of these patients. In our series, CT angiography was a safe, feasible, and useful tool in the management of VAD-related issues not detected by other available diagnostic modalities. As experience with this promising technique increases, CT angiography may become a primary tool in the evaluation of VAD patients.
1.Kannel WB: Incidence and epidemiology of heart failure. Heart Fail Rev 5: 167–173, 2000.
3.Rose EA, Gelijns AC, Moskowitz AJ, et al; Randomized Evaluation of Mechanical Assistance for the Treatment of Congestive Heart Failure (REMATCH) Study Group: Long-term mechanical left ventricular assistance for end-stage heart failure. N Engl J Med 345: 1435–1443, 2001.
4.Lietz K, Long JW, Kfoury AG, et al: Outcomes of left ventricular assist device implantation as destination therapy in the post-REMATCH era: Implications for patient selection. Circulation 116: 497–505, 2007.
5.Long JW, Healy AH, Rasmusson BY, et al: Improving outcomes with long-term “destination” therapy using left ventricular assist devices. J Thorac Cardiovasc Surg 135: 1353–1360, 2008.
6.Slaughter MS, Rogers JG, Milano CA, et al; HeartMate II Investigators: Advanced heart failure treated with continuous flow device. N Engl J Med 361: 2241–2251, 2009.
7.Horton SC, Khodaverdian R, Chatelain P, et al: Left ventricular assist device malfunction: An approach to diagnosis by echocardiography. J Am Coll Cardiol 45: 1435–1440, 2005.
8.Scalia GM, McCarthy PM, Savage RM, et al: Clinical utility of echocardiography in the management of implantable ventricular assist devices. J Am Soc Echocardiogr 13: 754–763, 2000.
9.Jaski BE, Miller DA, Hoagland PM, et al: Assessment of recurrent heart failure associated with left ventricular assist device dysfunction. J Heart Lung Transplant 24: 2060–2067, 2005.
10.Horton SC, Khodaverdian R, Powers A, et al: Left ventricular assist device malfunction: A systematic approach to diagnosis. J Am Coll Cardiol 43: 1574–1583, 2004.
11.Bryant AS, Holman WL, Nanda NC, et al: Native aortic valve insufficiency in patients with left ventricular assist devices. Ann Thoracic Surg 81: e6–e8, 2006.
12.Chumnanvej S, Wood MJ, MacGillivray TE, Melo MF: Perioperative echocardiographic examination for ventricular assist device implantation. Anesth Analg 105: 583–601, 2007.
13.Holman WL, Bourge RC, Fan P, et al: Influence of left ventricular assist on valvular regurgitation. Circulation 88: 309–318, 1993.
14.Baldwin RT, Duncan JM, Frazier OH, Wilansky S: Patent foramen ovale: A cause of hypoxemia in patients on left ventricular support. Ann Thorac Surg 52: 865–867, 1991.
15.Ferns J, Dowling R, Bhat G: Evaluation of a patient with left ventricular assist device dysfunction. ASAIO J 47: 696–698, 2001.
16.Parks JM, Nanda NC, Bourge RC, et al: Transesophageal echocardiographic evaluation of mechanical biventricular assist device. Echocardiography 7: 561–566, 1990.
17.Stainback RF, Croitoru M, Hernandez A, et al: Echocardiographic evaluation of the Jarvic 2000 axial-flow LVAD. Tex Heart Inst J 32: 263–270, 2005.
18.Catena E, Milazzo F, Montorsi E, et al: Left ventricular support by axial flow pump: The echocardiographic approach to device malfunction. J Am Soc Echocardiogr 18: 1422, 2005.
19.Akosah KO, Song A, Guearraty A, et al: Echocardiographic evaluation of patients with a left ventricular device. ASAIO J 44: M624–M627, 1998.
20.Sengupta PP, Khanderia BK: Transesophageal echocardiography. Heart 91: 541–547, 2005.
21.Horton SC, Khodaverdian R, Powers A, et al: Left ventricular assist device malfunction: A systematic approach to diagnosis. J Am Coll Cardiol 43: 1574–1583, 2004.
22.Jaski BE, Miller DA, Hoagland PM, et al: Assessment of recurrent heart failure associated with left ventricular assist device dysfunction. J Heart Lung Transplant 24: 2060–2067, 2005.
23.Smekal AV, Lachat ML, Willman JK, et al: Multislice spiral CT follow-up of a patient with implanted DeBakey ventricular assist device. Circulation 102: 1871–1872, 2000.
24.Jain VR, White CS, Pierson RN III: Imaging of left ventricular assist devices. J Thorac Imaging 20: 32–40, 2005.
25.Raman SV, Tran T, Simonetti OP, Sun B: Dynamic computed tomography to determine cardiac output in patients with left ventricular assist devices. J Thorac Cardiovasc Surg 137: 1213–1217, 2009.
26.Raman SV, Sahu A, Merchant AZ, et al: Noninvasive assessment of left ventricular assist devices with cardiovascular computed tomography and impact on management. J Heart Lung Transplant 29: 79–85, 2010.