Special Considerations for Durable Left Ventricular Assist Device Use in Small Patients

Large-scale clinical trials have shown improved outcomes as the durable mechanical circulatory support (MCS) field shifted to small, reliable continuous-flow blood pumps,^1–4 mirrored by real-world experience collected in registries.^5–8 Statistical power is achieved in aggregate results, but discerning clinical outcome differences by subpopulations (e.g., patient age,^9,10 gender,^11–13 race,⁸ or size^14,15) and optimizing therapy requires further examination.

Therapy application is closely controlled during clinical trials and assumed to be uniform irrespective of patient age, gender, race, size, etc. Device manufacturers, physicians, and patients share incentives to adhere to prescribed trial protocols, and inclusion and exclusion criteria preclude confounding factors from spuriously influencing outcomes. For example, in the MOMENTUM 3,¹⁶ ENDURANCE,¹⁷ and ENDURANCE Supplemental trials,¹⁸ body surface area (BSA) ≥1.2 m² was prescribed, thus nominally decoupling smaller patients from trial results. The average BSA during these trials was >2 m², suggesting that the study population was skewed to larger patients.

However, many therapy application steps cannot be practically prescribed, thus the vagaries of institutional history, experience, practices, physician convictions, and non-health-related patient demographics are introduced into trials, unrecorded. Furthermore, once devices attain regulatory agency approval, greater variation in patient inclusion and medical practice is realized.

In this article, special medical practice considerations for small body habitus patients are addressed by compiling critical themes established by panels of 29 physicians (including co-moderators) representing 11 countries (12× United States, 5× Japan, 3× Canada, 2× Germany, Belgium, Italy, Saudi Arabia, Singapore, Spain, Turkey, United Kingdom) during four advisory meetings, summarized by the meeting co-moderators (authors). Meeting details are provided in the Supplemental Digital Content, https://links.lww.com/ASAIO/A804.

Technological advances have enabled durable MCS device size reduction, limited by capacity requirements. Although successful partial MCS support has been demonstrated, full capacity (>5 L/min) is required to support the average adult chronic heart failure patient, and the research trend towards physiologically responsive MCS systems capable of increasing support during exercise discourages curtailing capacity. Small patients achieve target cardiac indices at lower volume flow rates, conceivably addressed by a smaller device, but the incremental benefit of a subtly smaller option might not outweigh the obstacles to bringing a second device to market. Therefore, optimizing the use of available one-size-fits-all commercial devices is a rational endeavor.

When implanting a device designed to fit most adults, greater challenge for smaller adult or pediatric patients is expected. Smaller patients may encounter difficulty managing relatively larger equipment in improperly sized accessories. Heterogeneous perceptions about MCS candidacy of small patients can lead to general healthcare inequity, including gender disparity (since women are on average smaller than men). Although potentially due to additional factors beyond size, the latter has been demonstrated by most multicenter LVAD analyses, in which fewer than 25% of overall LVAD patients are women.^5,12 Geographic and ethnic predisposition for small body habitus can magnify the importance of special therapy considerations, as for implant technique and post-operative care modifications in Japan.⁸ Optimally the goal is balancing real or perceived risks faced by small patients against unnecessarily denying them treatment. Determining small-patient factors that require enhanced attention and pooling experiences to refine practices were motivations for the advisory panels, and the output is described in this article.

Patient Candidacy and Selection

Among panelists, there was a perception disparity in what constitutes “smallness.” For example, the average historical adult male LVAD patient BSA is smaller in Japan than in the United States. Predispositions about size for candidacy could lead to exclusion from MCS consideration. American females often considered too small for implantation have a larger average BSA than Japanese men who are implanted. Physicians who treat pediatric patients are accustomed by necessity to using MCS devices designed for adults, sometimes in children with BSA below 1.0 m².⁹

Some panelists mentioned that experience has led to downward revising their viewpoints of appropriate BSA for most patients to 1.2 m² or lower. And although BSA is ubiquitous in assessing body size, it is vital to additionally regard other measures, such as heart size and thoracic cavity width and length. Panelists remarked that left ventricular end diastolic diameter (LVEDD) is relevant in determining MCS candidacy for two reasons: 1) A large LVEDD (>60 mm) has already generated thoracic adaptation that has partially created space for an LVAD, especially if the device immediately offloads and reduces the size of the ventricle; 2) A large LVEDD implies a dilated cardiomyopathic etiology, whereas a small LVEDD may suggest a restrictive or hypertrophic etiology, with potentially decreased LV filling and lower LVAD flows. LVEDD in HF is only weakly associated with BSA, driven more by etiology of heart failure than patient size, and even patients less than 8 years old can have an LVEDD >6 cm. Panelists agreed that small BSA patients with small LVEDD would likely not benefit from LVAD therapy with a ventricular inflow. Conversely, patients with small BSA but large LVEDD merit consideration for MCS and might be overlooked if considering only BSA.

Panel surgeons agreed that presurgical planning is more critical for small patients, as smaller chests generally offer fewer improvisation opportunities. Some surgeons employ advanced technologies such as computed tomography or magnetic resonance imaging-based digital (on the computer screen) or virtual reality (no screen, headset only) simulation to individualize surgical planning via patient images and digitized device solid models (Figure 1). The technology enables planning optimal device positioning in a patient’s extracardiac anatomy while accounting for apex-to-mitral distance and alignment of the LVAD inflow cannula with the mitral valve.^19,20

Figure 1.: Computed tomography- or magnetic resonance imaging-based patient images and digitized device solid models have been employed for virtual reality simulations to individualize surgical planning (provided by Cincinnati Children’s Heart Institute Digital, Media, and 3D-Modeling Division)

Cardiologists emphasized the importance of understanding underlying heart failure etiology and comorbidities, noting factors commonly associated with smaller patients such as frailty and cachexia that may adversely affect outcomes, including a sense that early right-heart failure risk and right ventricular assist device (RVAD) use may be aggravated by small adult patient size, encouraging further study.

Surgical Techniques and Device Implantation

Small thoracic size amplifies the criticality of anatomic fit and cannula position, including techniques to minimize pain. Presurgical planning does not replace perisurgical adaptation for internal organ shift upon ventricular decompression, and special attention to the plane of the attached apical cuff and the corresponding direction of the device when attached is required.

Although sternal-sparing approaches are common, some panelists prefer sternotomy to thoracotomy in small patients for wider access and visualization and may implant the device intra- or extrapericardially. Some surgeons described a preference to enter the pleural space to ensure adequate space for the device and to avoid shortening the long axis of the heart and interfering with right-heart function; some do this for all patients, small and large, others only when deemed necessary. All agreed that BSA <1.2 m² indicates the device should be positioned in the extrapericardial space. Surgeons with experience placing this device in ~25 kg children have avoided atelectasis by ensuring the device lays inferior to the left lower lobe and lateral and posterior to the dome of the diaphragm. To increase space for the device in the left thorax of patients less than 35 kg, one can dissect the anterior attachments of the left diaphragm and chest wall, allowing the costal margin and inferior ribs to move slightly anteriorly, increasing the anterior-posterior dimension.

Most concomitant factors, such as valvular pathology, are not addressed differently due to patient size. Extra driveline length realized in small patients often is compensated either with a contralateral abdominal counter-incision and bend or a gentle loop around the pump of a radius large enough to avoid subsequent driveline stress and fatigue. In patients with BSA <1.2 m², this loop can be positioned in a small, supradiaphragmatic pocket created to the right of the xyphoid process. To allay concern that driveline connector weight may confound immobilization and healing, a surgical technique was recommended, as described in the Supplemental Digital Content, https://links.lww.com/ASAIO/A804.

A small chest merits increased attention to prophylactic pain management. Contact between the device and ribs is more likely in smaller patients, and the risk of erosion, bleeding, and pain can be mitigated by fashioning a “pillow,” placed between the device and chest wall, made of GORE-TEX (W.L. Gore and Associates, Newark, DE) membrane and GELFOAM (Pfizer, New York, New York) sheet (Figure 2), palliating postoperative discomfort and facilitating easier device extraction during transplantation and device exchange. Extrapericardially implanted pumps should be covered, commonly with 0.1 mm GORE-TEX membrane, to prevent lung adhesions.

Figure 2.: A “pillow” can be fabricated with 0.1 mm GORE-TEX membrane sandwiching a GELFOAM sheet (above) to be placed between the device and the patient’s ribs (below) as a prophylactic maneuver to avert postsurgical pain (provided by Dr. Morales)

Surgeons who routinely implant the HeartMate 3 Left Ventricular Assist Device (HeartMate 3; Abbott, Chicago, Illinois) in patients <60 kg suggested attaching the bend relief only after the pump and graft have been lowered into the chest, and for patients <40 kg even the outflow graft connection to the pump may be withheld until the pump is safely inserted into the apical ring and lowered into the chest.

Maintaining transesophageal echocardiography (TEE) until the sternum is closed is recommended since closure of a small chest cavity can cause device shift and potentially affect flow due to a change in intraventricular septum position. TEE visualization guides device speed or position adjustment, often required in small patients, facilitating transfer to the cardiac intensive care unit with desirable hemodynamics, potentially preventing undesirable stress on the right ventricle.

Postsurgical Patient Management

Some panelists indicated that they initially start with lower device speeds, but a normal cardiac index, with smaller patients. Therapy titration by speed adjustment is based upon adequately decompressing the left ventricle without adversely affecting right-heart function; for most, this approach is no different for small patients.

Most panelists approach blood pressure measurement and target no differently for small adult patients than for others. For pediatric patients, blood pressure goals are determined by patient size and age. The consensus is that the lower cardiac output requirement of a small patient, especially if elderly, makes hypertension-related device flow reduction and potential device low-flow alarms more likely. For the HeartMate 3, a controller with an alarm threshold reduced from 2.5 to 2.0 L/min is available by physician request. Additionally, the tradeoff between hypertension-related risks (e.g., hemorrhagic stroke) and complications of excessive blood pressure reduction (e.g., presyncope and nausea) should be applied with differences among MCS devices in mind. An illustration of how running a device at a higher speed can be advantageous if safe to do so is in the Supplemental Digital Content, https://links.lww.com/ASAIO/A804.

Panelists indicated that anticoagulation strategy is no different for small patients, although for very small children aspirin dosing may be weight-based.

Summary

Preconceptions that vary by geography and experience lead to heterogeneous MCS therapy application based upon small size, with potential gender and geographical disparities. The criticality of factors other than BSA, such as LVEDD, may render the “eyeball test” of a patient’s smallness too crude to determine candidacy. Indeed, baseline heart size is a significant factor, and the small patient with large LVEDD may be appropriately qualified for MCS therapy yet overlooked on the basis of small BSA. Presurgical planning is more critical, as is anticipation of factors to which small patients may be more predisposed, such as postsurgical pain and lower LVAD flow. Although potential complications for small and large patients are similar, awareness of rate differences enhances patient management strategy and earlier detection and resolution of problems. Applying these considerations could result in expanding life-saving therapy to patients presently disqualified for their small size, decreasing existing gender and ethnic disparities, enhancing the quality of therapy, and building upon the existing experience that indicates no morbidity, mortality, or quality of life inferiority relative to larger patients, while promoting continuous future improvements (e.g., algorithms that appropriately titrate therapy) and trials.