- Email: [email protected]
Right Ventricular Dysfunction Is Associated with Gastrointestinal Bleeding in Patients Supported with Continuous-Flow LVADs
S12
The Journal of Heart and Lung Transplantation, Vol 32, No 4S, April 2014
8( ) Right Ventricular Dysfunction Is Associated with Gastrointestinal Bleeding in Patients Supported with Continuous-Flow LVADs C.T. Sparrow , M.E. Nassif, D.S. Raymer, E. Novak, S.J. LaRue, J.D. Schilling. Washington University School of Medicine, St. Louis, MO. Purpose: Gastrointestinal bleeding (GIB) is a significant complication in patients supported with continuous-flow LVADs. The impact of right ventricular (RV) dysfunction on the risk of GIB has not been investigated. This study tests the hypothesis that severe RV dysfunction is associated with an increased risk of GI bleeding post-LVAD. Methods: We retrospectively identified 219 patients who survived index hospitalization after implantation of Heart Mate II or Heartware HVAD from 6/2009 to 4/2013. Of these, 212 had a pre-LVAD echocardiogram in our system. Patients with severe RV dysfunction on pre-LVAD echocardiogram (n = 37) were compared to patients without severe RV dysfunction (n = 175). The primary outcome was time to GI bleed. Kaplan-Meier curves were created to evaluate time to first GI bleed and the log-rank test was used to compare event-free survival rates. Results: The majority of patients were male (78%) with a median INTERMACS profile of 2 at LVAD implant. There were no significant differences between cohorts with respect to demographics, comorbidities, device type, INR or aspirin strategy following LVAD implant. During follow up 80 patients had GIB events: 23 of 37 (62%) in the severe RV dysfunction group vs. 57 of 175 (33%) in the control group (p = 0.001). Kaplan-Meier analysis was also statistically significant, p = 0.02 (Figure). After adjustment for age and ICM, severe RV dysfunction was associated with increased risk of GIB (HR 1.859, 95% CI 1.143-3.024; p = 0.013). Conclusion: In this relatively large, single center sample of patients supported with continuous-flow LVADs, severe RV dysfunction on pre-LVAD echocardiogram was associated with an increased risk of GIB. Further studies are needed to investigate possible mechanisms by which RV dysfunction increases the risk of GIB and to identify patient populations who may benefit from alterations in anti-platelet and anticoagulation strategies.
9( ) Continuous Left Atrial Pressure Monitoring Improves Outcomes in Heartmate II Patients I. Gosev , R.C. Neely, M. Leacche, S. McGurk, Q. Javed, L. Cohn, G.S. Couper. Cardiac Surgery, Brigham and Women's Hospital, Jamaica Plain, MA. Purpose: Left atrial (LA) pressure is indicative of left ventricular(LV) function, LV preload and LV afterload. Current monitoring methods such as pulmonary artery lines give an indirect measurement of LV preload. We hypothesize that direct measurement of LV preload with a Codman Microsensor(off label) will facilitate more accurate fluid management and optimal ventricular assist device (VAD) settings, thereby improving outcomes after VAD implantation. Methods: Between 2/10-9/13, 98 patients underwent LVAD implantation (Heartmate II). 78.6%(77/98) had a Codman Microsensor placed in the LA for intra-op and post op monitoring(LA line group); 21.4%(21/98) had
conventional monitoring (no LA line group). Indications for VAD in LA line and no LA line groups differed for bridge to transplantation 67.5%(52/77) v. 38.1%(8/21)(p< 0.022), destination therapy 29.9%(23/77) v. 57.7%(12/21) (p< 0.004), respectively. No differences were seen for bridge to recovery in LA line 1.3%(1/77) v. no LA line 4.8%(1/21), p< 0.384. Prior VAD malfunction was seen in 1.3%(1/77) in LA line. Results: Mean age was 54.1 yrs (+/-11.7). Median LV ejection fraction 15%(15, 20), pre VAD pulmonary capillary wedge pressure 23 mmHg (18,27), and baseline demographics were similar between groups, except mean body surface area in LA line 2.01(+/-.24)m2 v. no LA line 1.78(+/-.20) m2(p< 0.001) and women in LA line 18.2%(14/77) v. 42.9%(9/21), p< 0.038. Median perfusion time was shorter in LA line 106 mins (82,151) v. no LA line 154 mins (111, 190), p< 0.005. The LA line group had significantly shorter median ventilation time 17 hrs (10, 88) v. 61 hrs (13,234)(p< 0.033), initial ICU time 153 hrs (113, 253) v. 225 hrs (145, 356)(p< 0.022), and total ICU time of 154 hrs(115,258) v. 221 hrs(162,385), p< 0.019. Median post operative length of stay trended towards significance in LA line 22 days (16,39) v. no LA line 31 days (22,40), p< 0.053. Overall operative mortality was 5.1%(5/98) and 3.9%(3/77) in LA line v. 9.5%(2/21) no LA line, p< 0.291. Conclusion: LA pressure monitoring system using a Codman Microsensor offers safe, reliable and accurate monitoring for LV preload during and after LVAD implantation. We observed shorter ventilation time, ICU and hospital length of stay in continuously monitored patients compared to conventional monitoring. Continuous LA pressure monitoring technique may improve outcomes after LVAD placement. 1( 0) Relationship Between Speed and Left Ventricular Size in HVAD Patients: Lessons from the HVAD Ramp Test Study N. Uriel , A.P. Levin, K.P. Mody, M. Dionizovik-Dimanovski, A.R. Garan, M. Yuzefpolskaya, H. Takayama, P.C. Colombo, Y. Naka, U.P. Jorde. Medicine, Columbia University, New York, NY. Purpose: Ramp tests have been widely used both for speed adjustment and device malfunction identification in axial flow LVAD pts (HM2). The aim of this study was to test for the first time the ramp test in centrifugal flow HVAD pts. Methods: A prospective ramp test protocol for HVAD pts was conducted in 15 consecutive pts prior to discharge, or at the time of suspected device malfunction. Vital signs, device and echo parameters, including left ventricular end-diastolic dimension (LVEDD), frequency of aortic valve (AV) opening and valvular insufficiency, were recorded following 2 minute increments of 100 rpm from 2,300 to 3,200 rpm. Results: Eighteen ramp tests were performed, 13 for speed optimization and 5 for assessment of device malfunction. Mean speed was 2642±245 rpm, and the AV closed at a mean speed of 2761±225 rpm, with the exception of 2 pts whose AV never closed. A nonlinear relationship between device speed and LVEDD was observed in pts with normally functioning HVADs. The LVEDD did not change significantly before the AV was closed, but rapidly decompressed after, with mean LVEDD slopes of -0.13±0.12 before and -0.26±0.22 after AV closure (p= 0.076). Device flow rapidly increased prior to AV closure with a mean slope of 0.25±0.09, and flattened after it, with a mean slope of 0.06±0.19 (p= 0.014). Two pts had device obstruction with completely flat LVEDD slopes and erratic flow data. Fifteen tests were conducted to completion, without any events. Three were stopped early, one due to VT and two because of suction events. Conclusion: This is a novel study investigating the relationship between speed and LVAD decompression in the HVAD pts. The HVAD decompresses the LV in a nonlinear way, with significantly less decompression than seen historically HM2. Thus, LVEDD slopes from HVAD ramp tests should not be analyzed based on the HM2 data available. Speed adjustment in HVAD patients should not be based on the degree of LV decompression. Flow slopes may offer a better approach to assess device malfunction in HVAD patients.
The Journal of Heart and Lung Transplantation, Vol 32, No 4S, April 2014
8( ) Right Ventricular Dysfunction Is Associated with Gastrointestinal Bleeding in Patients Supported with Continuous-Flow LVADs C.T. Sparrow , M.E. Nassif, D.S. Raymer, E. Novak, S.J. LaRue, J.D. Schilling. Washington University School of Medicine, St. Louis, MO. Purpose: Gastrointestinal bleeding (GIB) is a significant complication in patients supported with continuous-flow LVADs. The impact of right ventricular (RV) dysfunction on the risk of GIB has not been investigated. This study tests the hypothesis that severe RV dysfunction is associated with an increased risk of GI bleeding post-LVAD. Methods: We retrospectively identified 219 patients who survived index hospitalization after implantation of Heart Mate II or Heartware HVAD from 6/2009 to 4/2013. Of these, 212 had a pre-LVAD echocardiogram in our system. Patients with severe RV dysfunction on pre-LVAD echocardiogram (n = 37) were compared to patients without severe RV dysfunction (n = 175). The primary outcome was time to GI bleed. Kaplan-Meier curves were created to evaluate time to first GI bleed and the log-rank test was used to compare event-free survival rates. Results: The majority of patients were male (78%) with a median INTERMACS profile of 2 at LVAD implant. There were no significant differences between cohorts with respect to demographics, comorbidities, device type, INR or aspirin strategy following LVAD implant. During follow up 80 patients had GIB events: 23 of 37 (62%) in the severe RV dysfunction group vs. 57 of 175 (33%) in the control group (p = 0.001). Kaplan-Meier analysis was also statistically significant, p = 0.02 (Figure). After adjustment for age and ICM, severe RV dysfunction was associated with increased risk of GIB (HR 1.859, 95% CI 1.143-3.024; p = 0.013). Conclusion: In this relatively large, single center sample of patients supported with continuous-flow LVADs, severe RV dysfunction on pre-LVAD echocardiogram was associated with an increased risk of GIB. Further studies are needed to investigate possible mechanisms by which RV dysfunction increases the risk of GIB and to identify patient populations who may benefit from alterations in anti-platelet and anticoagulation strategies.
9( ) Continuous Left Atrial Pressure Monitoring Improves Outcomes in Heartmate II Patients I. Gosev , R.C. Neely, M. Leacche, S. McGurk, Q. Javed, L. Cohn, G.S. Couper. Cardiac Surgery, Brigham and Women's Hospital, Jamaica Plain, MA. Purpose: Left atrial (LA) pressure is indicative of left ventricular(LV) function, LV preload and LV afterload. Current monitoring methods such as pulmonary artery lines give an indirect measurement of LV preload. We hypothesize that direct measurement of LV preload with a Codman Microsensor(off label) will facilitate more accurate fluid management and optimal ventricular assist device (VAD) settings, thereby improving outcomes after VAD implantation. Methods: Between 2/10-9/13, 98 patients underwent LVAD implantation (Heartmate II). 78.6%(77/98) had a Codman Microsensor placed in the LA for intra-op and post op monitoring(LA line group); 21.4%(21/98) had
conventional monitoring (no LA line group). Indications for VAD in LA line and no LA line groups differed for bridge to transplantation 67.5%(52/77) v. 38.1%(8/21)(p< 0.022), destination therapy 29.9%(23/77) v. 57.7%(12/21) (p< 0.004), respectively. No differences were seen for bridge to recovery in LA line 1.3%(1/77) v. no LA line 4.8%(1/21), p< 0.384. Prior VAD malfunction was seen in 1.3%(1/77) in LA line. Results: Mean age was 54.1 yrs (+/-11.7). Median LV ejection fraction 15%(15, 20), pre VAD pulmonary capillary wedge pressure 23 mmHg (18,27), and baseline demographics were similar between groups, except mean body surface area in LA line 2.01(+/-.24)m2 v. no LA line 1.78(+/-.20) m2(p< 0.001) and women in LA line 18.2%(14/77) v. 42.9%(9/21), p< 0.038. Median perfusion time was shorter in LA line 106 mins (82,151) v. no LA line 154 mins (111, 190), p< 0.005. The LA line group had significantly shorter median ventilation time 17 hrs (10, 88) v. 61 hrs (13,234)(p< 0.033), initial ICU time 153 hrs (113, 253) v. 225 hrs (145, 356)(p< 0.022), and total ICU time of 154 hrs(115,258) v. 221 hrs(162,385), p< 0.019. Median post operative length of stay trended towards significance in LA line 22 days (16,39) v. no LA line 31 days (22,40), p< 0.053. Overall operative mortality was 5.1%(5/98) and 3.9%(3/77) in LA line v. 9.5%(2/21) no LA line, p< 0.291. Conclusion: LA pressure monitoring system using a Codman Microsensor offers safe, reliable and accurate monitoring for LV preload during and after LVAD implantation. We observed shorter ventilation time, ICU and hospital length of stay in continuously monitored patients compared to conventional monitoring. Continuous LA pressure monitoring technique may improve outcomes after LVAD placement. 1( 0) Relationship Between Speed and Left Ventricular Size in HVAD Patients: Lessons from the HVAD Ramp Test Study N. Uriel , A.P. Levin, K.P. Mody, M. Dionizovik-Dimanovski, A.R. Garan, M. Yuzefpolskaya, H. Takayama, P.C. Colombo, Y. Naka, U.P. Jorde. Medicine, Columbia University, New York, NY. Purpose: Ramp tests have been widely used both for speed adjustment and device malfunction identification in axial flow LVAD pts (HM2). The aim of this study was to test for the first time the ramp test in centrifugal flow HVAD pts. Methods: A prospective ramp test protocol for HVAD pts was conducted in 15 consecutive pts prior to discharge, or at the time of suspected device malfunction. Vital signs, device and echo parameters, including left ventricular end-diastolic dimension (LVEDD), frequency of aortic valve (AV) opening and valvular insufficiency, were recorded following 2 minute increments of 100 rpm from 2,300 to 3,200 rpm. Results: Eighteen ramp tests were performed, 13 for speed optimization and 5 for assessment of device malfunction. Mean speed was 2642±245 rpm, and the AV closed at a mean speed of 2761±225 rpm, with the exception of 2 pts whose AV never closed. A nonlinear relationship between device speed and LVEDD was observed in pts with normally functioning HVADs. The LVEDD did not change significantly before the AV was closed, but rapidly decompressed after, with mean LVEDD slopes of -0.13±0.12 before and -0.26±0.22 after AV closure (p= 0.076). Device flow rapidly increased prior to AV closure with a mean slope of 0.25±0.09, and flattened after it, with a mean slope of 0.06±0.19 (p= 0.014). Two pts had device obstruction with completely flat LVEDD slopes and erratic flow data. Fifteen tests were conducted to completion, without any events. Three were stopped early, one due to VT and two because of suction events. Conclusion: This is a novel study investigating the relationship between speed and LVAD decompression in the HVAD pts. The HVAD decompresses the LV in a nonlinear way, with significantly less decompression than seen historically HM2. Thus, LVEDD slopes from HVAD ramp tests should not be analyzed based on the HM2 data available. Speed adjustment in HVAD patients should not be based on the degree of LV decompression. Flow slopes may offer a better approach to assess device malfunction in HVAD patients.