Search

BACKGROUND: Gastrointestinal bleeding (GIB) is a common complication of left ventricular assist device (LVAD) therapy accounting for frequent hospitalizations and high resource utilization. METHODS: We previously developed an endoscopic algorithm emphasizing upfront evaluation of the small bowel and minimizing low-yield procedures in LVAD recipients with GIB. We compared the diagnostic and therapeutic yield of endoscopy, health-care costs, and re-bleeding rates between conventional GIB management and our algorithm using chi-square, Fisher’s exact test, Wilcoxon-Mann-Whitney, and Kaplan-Meier analysis. RESULTS: We identified 33 LVAD patients with GIB. Presentation was consistent with upper GIB in 20 (61%), lower GIB in 5 (15%), and occult GIB in 8 (24%) patients. Forty-one endoscopies localized a source in 23 (56%), resulting in 14 (34%) interventions. Algorithm implementation compared with our conventional cohort was associated with a 68% increase in endoscopic diagnostic yield (P< .01), a 113% increase in therapeutic yield (P= .01), a 27% reduction in the number of procedures per patient (P < .01), a 33% decrease in length of stay (P < .01), and an 18% reduction in estimated costs (P < .01). The same median number of red blood cell transfusions were used in the 2 cohorts, with no increase in re-bleeding events in the algorithm cohort (33.3%) compared with our conventional cohort (43.7%). CONCLUSIONS: Our endoscopic management algorithm for GIB in LVAD patients proved effective in reducing low-yield procedures, improving the diagnostic and therapeutic yield of endoscopy, and decreasing health-care resource utilization and costs, while not increasing the risk of a re-bleeding event.

Purpose Driveline infection (DLI) is a major cause of morbidity and mortality of LVAD therapy. Herein, we aimed to study: 1) whether temporal changes in DL care and pump type had influenced the rate of DLIs; and 2) the long-term outcome after DLI diagnosis based on treatment strategy. Methods Pts implanted with LVADs from 2/09 to 5/19 were included. The cohort was divided into 4 eras (Es) based on DL care strategy: E1: original care (1/09-3/11); E2: standard dressing kit (4/11-3/16); E3: marking of DL exit site prior to implant (4/16-6/17); E4: strict no shower policy (7/17-). The primary endpoint was freedom from DLI by era and pump type. Survival after DLI diagnosis based on treatment strategy was also investigated. Cox proportional hazard model for the primary endpoint was fit using the following variables: age, sex, BMI, etiology, DM, baseline creatinine, implant strategy, INTERMACS and pump type. Results 593 pts, age 56±14y; 20% F; 371(63%) HMII, 160(27%) HM3, 62 (10%) HVAD were studied. DLI occurred in 85(14.3%) pts at a median [IQR] of 357 [198-532] d post LVAD. Baseline characteristics were similar in pts with and without DLI. E1 was associated with the highest incidence of DLI (Fig.1a). 1-y freedom from DLI was similar among pumps: 90.1% HMII, 87.5% HM3 and 92.5% HVAD (p=0.86). 96 organisms were cultured in 85 DLI pts, the 3 most common were: MSSA 34/96 (35%), P. Aeruginosa 17/96(18%), S. marcescens 10/96(10%). 58(68%) of the DLI pts were treated with antibiotics alone and 27(32%) required additional surgical debridement (SD) at a median of 42 (2-225) d after DLI diagnosis. Survival after DLI and after SD are shown in Fig 1b-c. At the end of follow up, out of 85 pts with DLI, 15(18%) pts remained on support, 40(47%) were transplanted and 27(32%) died. At the end of follow up, out of 27 post-SD pts, 3pts remained on support, 11 were transplanted and 13 died. Conclusion Temporal changes in DLI care reduced the rates of DLI. Survival in pts who develop DLI is reduced, particularly in those who require surgical intervention.

Purpose Lack of nocturnal blood pressure (BP) reduction (BP dipping) has been associated with cerebrovascular and cardiovascular events in the general population. However, the effect of a deranged circadian BP variability between day and night has never been studied in patients on left ventricular assist device (LVAD) support, where stroke still represents a major complication. We hypothesized that decreased BP dipping significantly increases the risk of stroke in patients on LVAD support. Methods HeartMate II (HMII) outpatients were prospectively studied with an ambulatory BP monitor (Mobil-O-Graph) previously validated in this patient population. Systolic BP (SBP), diastolic BP (DBP) and mean arterial pressure (MAP) were recorded every 30 minutes during the day and every hour during the night for 24 hours. Circadian patterns, expressed as the percentage of nocturnal SBP, DBP, and MAP reduction from daytime values, were assessed in Cox regression models with stroke as primary endpoint. Results 29 patients were studied (age 59 ±15 yrs, 76% male, 38% ischemic, 69% DT). 6 patients (20%) experienced a stroke 879±591 days post LVAD implant. Nocturnal reductions of SBP, DBP, and MAP were 5.0 ± 6.1%, 5.0 ± 5.6% and 5.0 ± 5.5%, respectively. Preserved nocturnal reductions of SBP (HR 0.790 [95% CI 0.658 - 0.949]), DBP (HR 0.707 [95% CI 0.559 - 0.894]) and MAP (HR 0.703 [95% CI 0.554 - 0.894]) were associated with a decreased risk of stroke. When patients were grouped by the median nocturnal reduction in MAP (5.4%), patients with reduced dipping had a significantly greater risk of stroke (Figure 1). Conclusion This study is the first to indicate that reduced BP dipping at night increases the risk of stroke in patients on HMII support. Further studies are warranted to investigate, mechanistically, the relationship between reduced BP dipping and sympathetic nervous system activation, and, clinically, its association with the risk of stroke in patients on HM3 support.

Purpose Declining renal function portends poor prognosis in advanced HF. Hemodynamic optimization with diuretics, inotropes, and temporary mechanical circulatory support is frequently performed for optimization of end-organ function before LVAD surgery. We aimed to study the relationship between long- and short-term changes in serum creatinine (sCr) that occur prior to implant and postoperative outcome in LVAD pts. Methods We retrospectively identified 146 LVAD pts with sCr at 3 time points: 1-year (1y), day of index admission, and 1-day (1d) prior to LVAD. Pts were classified based on the changes in sCr that occurred between 1y and 1d prior to LVAD as: Improved = sCr reduction >10% (n=42); Stable = sCr change ≤10% (n=50); Worsened = sCr increase >10% (n=54). Pts were also classified as Optimized using the following exploratory cutoffs: sCr reduction of ≥10, 15, and 20% during the index admission. The primary endpoint was survival free from renal replacement therapy (RRT) after LVAD. Multivariable Cox models were developed to adjust for age, sex, race and sCr 1d pre-LVAD. Results On average, sCr increased from 1 y to the time of index admission, while subsequently declined prior to LVAD (Fig A). Among 147 pts, 47 (31%) met the primary endpoint. In univariable model, Stable 1y-1d sCr was associated with a reduced risk for the primary endpoint (Fig B). After adjustment, there was only a trend towards significance in Stable vs. Improved (p=0.06). Optimization of sCr during index admission was not associated with improved outcomes. Furthermore, in an adjusted model, a decrease in sCr≥15% was significantly associated with increased risk for the primary endpoint (Fig B). Conclusion In the year preceding LVAD implant, a stable renal function is associated with a trend towards improved post-LVAD survival free from RRT. Moreover, a decrease in sCr≥15% during index admission was associated with worse outcome. The effect of sarcopenia on sCr levels in advanced HF may explain these paradoxical findings.

Ventricular arrhythmias (VAs) are common after implantation of a left ventricular assist device (LVAD) and in a subset of patients may be refractory to medication. Morbidity from VA in this population includes right ventricular failure (RVF). We sought to evaluate the efficacy of catheter ablation for VA in patients with LVAD. A retrospective analysis of patients supported by continuous-flow LVAD referred for catheter ablation of ventricular tachycardia (VT) between 2008 and the present was performed. Seven patients were referred for VT ablation an average of 236 ± 292 days after LVAD implantation. Three patients (42.9%) developed RVF in the setting of intractable arrhythmias. A transfemoral approach was used for six patients (85.7%) and an epicardial for one patient (14.3%). The clinical VT was inducible and successfully ablated in six patients (85.7%). The location of these arrhythmias was apical in three cases (42.9%). A total of 13 VTs were ablated in seven patients. Although the majority had reduction in VA frequency, recurrent VAs were observed in six patients (85.7%). One patient (14.3%) experienced a bleeding complication after the procedure. For patients with a high VA burden after LVAD implantation, VT ablation is safe and feasible, but VA frequently recurs.

Purpose In the normal, pulsatile circulation, the hemodynamic pattern of the middle cerebral artery (MCA) differs from the central retinal artery (CRA). Whether continuous flow from a left ventricular assist device (LVAD) alters this relationship is not known. Since serious adverse events in different end-organs, such as stroke and GI bleeding, are frequent in LVAD patients, comparing hemodynamics in different arterial beds may enhance our understanding of the specific pathophysiology of these complications. Hypothesis Because of continuous flow, hemodynamic patterns between the MCA and CRA are more similar in HeartMate II (HM II) patients compared with healthy individuals. Methods MCA and CRA time-averaged maximal flow velocity (TAMAX), pulsatility index (PI) and resistance index (RI) were determined using Doppler ultrasound with angle correction, in 20 HM II pts and 21 healthy individuals. To test the study hypothesis, data were statistically analysed for both group differences and group variances. Results Although PI and RI were markedly lower in HM II patients in both the MCA and CRA (P Conclusion The typical differences that are observed in small- and microcirculatory hemodynamics of healthy individuals are altered in HM II patients. Notably, the wide ranges of haemodynamics observed in the HM II group may carry important implications with respect to the pathophysiology of specific end-organ complications, and, eventually, the identification of high-risk patients.

Background On April 4, 2012, the U.S. Food and Drug Administration issued a Class 1R recall of the HeartMate II (Thoratec Corporation, Pleasanton, CA) left ventricular assist device (LVAD) due to spontaneous detachment of the bend relief from its intended position in patients implanted with the most recent version of the HM II. This study examined the incidence and timing of outflow graft bend relief disconnection in patients implanted with the HM II LVAD. Methods All patients supported with the modified version of the HM II LVAD were asked to report for dedicated abdominal X-ray imaging to assess the position of the bend relief. Also performed was a retrospective review of X-ray images of all patients who had previously been supported with this version but had since received a transplant, undergone LVAD explant, or died. Results Between March 9, 2011, and April 9, 2012, 59 patients underwent primary implant with the modified version HM II. Follow up X-ray images were available for 56 patients (95%). The bend relief was found fully disconnected in 6 of 56 (11%) and partially disconnected in 13 (23%). Two of 6 patients (33%) with full bend relief disconnection and 1 of 13 of the initially partially disconnected patients (7.7%) required urgent surgical intervention due to symptoms of hemolysis and/or heart failure. Conclusions Bend relief disconnection is common and may be observed immediately after implant but may also develop over time. Full bend relief disconnect may present with hemolysis and/or heart failure symptoms and often requires surgical revision. Surveillance abdominal X-ray imaging should be performed routinely on all patients who were implanted with the modified version HM II.

Purpose Ventricular arrhythmias (VA) are common in patients with a continuous flow left-ventricular assist device (CF-LVAD) but prospective data to support the routine ICD use in these patients are lacking. The aim of this study was to evaluate the prevalence and significance of Vas as well as the role of ICDs in this population. Methods and Materials All patients with a long-term CF-LVAD receiving care at our institution were enrolled. ICDs were interrogated at baseline and throughout prospective follow-up. VA was defined as VT/VF lasting > 30 seconds or effectively terminated by appropriate ICD therapy. The primary outcome was the occurrence of late Vas (more than 30 days after CF-LVAD implant); secondary outcomes were early Vas (within 30 days post-operatively) and all-cause mortality. Results Ninety-four patients were enrolled; seventy-seven had an ICD and 17 did not. Five patients with an ICD had it de-activated or a depleted battery not replaced during the study. Patients without an ICD were more likely to be women (58.8% vs. 90.9%, p=0.003), have shorter disease duration (35.3% vs. 97.4%, p + 1.0cm vs. 7.0 + 0.9cm, p Conclusions CF-LVAD patients with pre-operative Vas are at risk of recurrent Vas. Active ICD may minimize the morbidities associated with sustained Vas. Patients without pre-operative Vas are at low risk of post-operative Vas and may not need ICD therapy.

Purpose Anticoagulation therapy in Continuous Flow (CF) Left Ventricular Assist Device (LVAD) pts continues to be challenging due to high frequency of Gastrointestinal (GI) bleeding and device thrombosis. This study sought to determine if using a Home INR Machine (HINRM) will reduce the risk of these complications. Methods and Materials All pts monitored by HINRM during LVAD support were enrolled. Data on anticoagulation regimen and device related complications were collected. Results 14 CF-LVAD pts who have used or currently use HINRM were identified. Mean age was 68 ± 11 years old, 86% were male and 57% were destination therapy. 86% (n=12) were compliant and their INR was maintained within therapeutic range 67±24% of the time. The INR values were reported on average once every 9.2 ± 5.3 days. While not using HINRM (386±367 days), 43% (n=6) of these pts were readmitted for GI bleeding (4 pts), device thrombosis (1 pt) and epistaxis (1 pt). While using HINRM (376±250 days), 21% (n=3) of these pts were readmitted for 3 cases of GI bleeding, 1 case of epistaxis and none from DT. Two of the GI bleeding cases occurred during a 3-month period of non-compliance from a single pt. The total event rate was reduced from 0.41 to 0.28 events/person-year, respectively. Conclusions Our data suggests that self-monitoring with HINRM among CF-LVAD pts is a feasible form of maintaining adequate INR range with good compliance. Lower readmission rate for GI bleeding and device thrombosis was noted after using HINRM than prior. Larger cohorts with longer follow up times are warranted. [ figure 1 ]