- Email: [email protected]
Pocket Infection in Patients with Left Ventricular Assist Devices: An INTERMACS Analysis
S84
The Journal of Heart and Lung Transplantation, Vol 32, No 4S, April 2013 Baylor College of Medicine, Houston, TX; 3Pediatric Cardiothoracic Surgery, Cincinnati Children’s Hospital Medical Center, Cincinnati, OH; 4Pharmacy, Texas Children’s Hospital, Houston, TX; 5Department of Pediatrics, Section of Infectious Disease, Texas Children’s Hospital, Baylor College of Medicine, Houston, TX. Purpose: Infectious complications constitute a major cause of morbidity and mortality in adult patients supported with left ventricular assist devices (VADs). Incidence and patient outcomes related to infectious complications in pediatric patients on VAD support remain largely unknown. The aim of this study was to determine the incidence of infection among pediatric VAD recipients and to characterize the microbiology, associated risk factors and clinical outcome. Methods and Materials: We conducted a retrospective chart review of all patients undergoing VAD support for Z2 weeks at Texas Children’s Hospital from June 1999-December 2011. Infections were categorized as VAD-specific, VAD-related or non-VAD related using the International Society for Heart and Lung Transplantation (ISHLT) definitions for VAD infections. Results: Fifty-two VADs were implanted in 51 patients. Thirty-five (69%) patients had 92 infections on VAD support and included: 10 VAD-specific infections, 23 VAD-related infections, and 59 non-VAD infections. Overall rate of VAD-infections (specificþrelated) was 9.4/ 1000 days of VAD support. The most common pathogens were Staphylococcus aureus, coagulase negative staphylococci, Pseudomonas aeruginosa and Candida species. Three (37.5%) of 8 deaths occurring during VAD support were directly related to infections. Prolonged hospital stay (p¼0.005) and prior cardiac transplantation with rejection (p¼0.018) were significantly associated with development of VAD-infections. Conclusions: Infectious complications are common in pediatric patients undergoing VAD support. By identifying and controlling risk factors, it may be possible to reduce the occurrence of VAD infections in pediatric patients.
212 Electronic medical records were reviewed for infection details to supplement data available in our transplant databases. DLIs were defined according to Hannan, et al (2011). Results: From 7/04 to 7/11, 181 LVADs were implanted and 44 DLIs occurred during 304 person-years of LVAD support. Hazard for DLI after LVAD was constant at 2.2%/mo but increased to 10%/mo from 6-9 months. Pre-, peri-, post-operative variables did not correlate with risk for DLI. ‘‘Trauma’’ to drive line was described in 20/44 (45%) of DLI vs. 3/137 (2.1%) without DLI but this was not routinely documented. DLI doubled the risk for death before transplant (HR 2.16, CI95% 1.18, 3.96). Pseudomonas aeruginosa (PA) was the most common pathogen and often replaced initial pathogens over time. Superficial DLI progressed at a mean 173d (61-362d) to deep DLI in 9/39 (23%), 8/9 due to PA, 1/9 Staphylococcus aureus. Multidrug resistance emerged in 4 PA and 3 other gram negative DLIs. Other complications of antibiotics rarely occurred. Conclusions: DLI prevalence accumulates over time in patients on LVAD support. Our data demonstrate a peak hazard of DLI from 6-9 months after implantation. DLI is an independent predictor of death before transplantation in our population. PA, more than other organisms, progressed to deep DLI and became more drug resistant over time on support. 211 Infectious Complications and Outcomes in Children Supported with Left Ventricular Assist Devices A.G. Cabrera,1 M.S. Khan,2 D.L.S. Morales,3 D.W. Chen,2 B.S. Moffett,4 J.F. Price,1 W.J. Dreyer,1 S.W. Denfield,1 A. Jeewa,1 C.D.Fraser Jr.,2 J.G. Vallejo.5 1Department of Pediatrics, Section of Pediatric Cardiology, Texas Children’s Hospital, Baylor College of Medicine, Houston, TX; 2Michael E. DeBakey Department of Surgery, Division of Congenital Heart Surgery, Texas Children’s Hospital,
Thrombotic Complications Increase after Percutaneous Site/Pocket Infection in Patients with Left Ventricular Assist Devices: An INTERMACS Analysis M.R. Shah,1 D.C. Naftel,2 M.A. Miller,1 J.T. Baldwin,1 E.D. Feller,3 S.S. Desai,4 R. John,5 R.L. Kormos,6 W.L. Holman.2 1National Heart, Lung, and Blood Institute, Bethesda, MD; 2University of Alabama at Birmingham, Birmingham, AL; 3University of Maryland Medical Center, Baltimore, MD; 4Inova Fairfax Hospital, Falls Church, VA; 5 University of Minnesota, Minneapolis, MN; 6University of Pittsburgh Medical Center, Pittsburgh, PA. Purpose: Previous studies suggest infection may confer risk for thrombotic complications (TC) after LVAD. We hypothesized that in discharged (disch) LVAD patients (pts), there is an increased rate of TC within 6 weeks (wks) after a major infection. Methods and Materials: We reviewed INTERMACS (4/2008 to 6/2012) to identify disch pts with continuous flow LVAD, with Z1 infection after disch. INTERMACS definitions were used. Infection was categorized as: 1) pump infection (pump); 2) percutaneous site/pocket
Percutaneous Site/Pocket Infection
Total TC Neurological dysfunction Pump thrombosis Non-CNS TC Hemolysis
Pre-infection (n; rate)
Post-infection (n; rate) p-value
12; 1.8 6; 0.9
33; 4.9 18; 2.7
0.005 0.01
2; 0.3
7; 1.0
0.06
0; 0.0 4; 0.6
1; 0.2 7; 1.04
– 0.48
Abstracts
S85
infection (device related); 3) sepsis; and, 4) localized non-device infection (non-device). TC were: 1) neurological dysfunction; 2) pump thrombosis; 3) arterial non-central nervous system and, 4) hemolysis (plasma free hemoglobin 440 mg/L). TC rates (TC per 100 mo of pt exposure) 6 wks pre-infection were compared to rates 6 wks postinfection. P-values were calculated with McNemar’s test. Results: Of 4124 disch LVAD pts, 1253 (30%) had a major infection post-disch (pump, n¼5; device related, n¼517; sepsis, n¼203; nondevice, n¼528). Key features of infected pts (mean⫾SD and %) were: age (54.6⫾8.3 yrs.), male (75.4%), diabetes (43.8%), destination therapy (27.9%). Last INR prior to TC was 2.1⫾1.3. There was no significant difference in TC rate after an infection vs pre-infection (6.63 vs 6.27; p¼0.67). When analyzed by category of infection, there was a higher rate of TC after device related infection. Conclusions: In disch LVAD pts there was a significantly higher rate of TC driven by neurological dysfunction and pump thrombosis within 6 wks of a device related infection. There was no significant association between other infections and TC. This data suggests increased risk of TC following device related infection. 213 Incidence and Characterization of Percutaneous Lead Damage in the Heartmate II Left Ventricular Assist Device D. Kalavrouziotis,1 A. Massiello,2 M.Z. Tong,1 R.C. Starling,3 R. Fryc,4 G. Heatley,4 D.J. Farrar,4 N. Moazami.1 1Department of Thoracic and Cardiovascular Surgery, Cleveland Clinic, Cleveland, OH; 2Department of Biomedical Engineering, Lerner Research Institute, Cleveland Clinic, Cleveland, OH; 3Department of Cardiovascular Medicine, Cleveland Clinic, Cleveland, OH; 4Thoratec Corporation, Pleasanton, CA. Purpose: The incidence of percutaneous lead failure due to wear and traumatic fracture among patients supported with a Heartmate (HM) II left ventricular assist device (LVAD) has not been previously reported. Methods and Materials: The study is a retrospective review of all HM II LVAD percutaneous driveline malfunctions reported to Thoratec Corporation. Categorization of wear and failure by location and severity was performed and the incidence of adverse clinical outcomes as well as the management of the dysfunctional leads were examined. Results: Between 2004 and October 2012, 12,969 HM II pumps were implanted worldwide. The incidence of percutaneous lead malfunction was 1,418 events occurring in 1,198 pumps (9.2%) over a cumulative support period of 13,932 patient-years (maximum ¼ 7.3 years). Reported lead damage was mostly in the externalized part of the cable (87.2%). Lead malfunction was managed with external reinforcement in 7.5% of the implanted pumps, while mortality or significant morbidity including pump exchange or urgent transplant, or more complex endcable replacement occurred in 2.3%. Percutaneous lead malfunction not associated with adverse events was managed by clamshell reinforcement of the connector strain relief or by tape/silicone cable reinforcement. The cumulative incidence of internal lead failures has decreased with two lead design revisions: 1.6% in 2004 (original design) vs. 0.01% in 2010 (redesigned internal pump end bend relief). Conclusions: The incidence of HM II percutaneous lead malfunction associated with serious adverse events is low and the majority of patients who experience lead damage have been successfully managed by external repairs. The incidence of both internal and external lead failures has diminished since 2004 with progressive iterations and improvements in lead design. 214 Predisposition to Infection and Bleeding in Mechanical Assist Devices (LVADs) Recepients after Dental Procedures: A Single Center Experience K.C. McCants, J.B. McCants, P.S. Combs, P. Raheja, M.S. Slaughter, E. Birks. Advanced Heart Failure and Transplant, University of Louisville & Jewish Hospital, Louisville, KY. Purpose: Currently more than 7,000 patients with heart failure are supported with LVADs. Few patients with LVADs undergo dental
procedures and are at risk of bleeding due to their anticoagulation/ antiplatelet medication profile and acquired VWF deficiency. Many centers follow the AHA guidelines for the prophylaxis of bacterial endocarditis in patients with prosthetic devices. Though these guidelines are comprehensive, they do not specifically address prophylactic recommendations for patients with LVADs. We would assume that these guidelines hold true for LVADs; however there is no current evidence. Methods and Materials: To assess the risk of infections and bleeding complications during dental procedures in patients with LVADs, we designed a protocol based pilot study of LVAD patients using the AHA guidelines. We followed 19 patients respective clinical course for administration of preprocedural antibiotics, continuation of their
The Journal of Heart and Lung Transplantation, Vol 32, No 4S, April 2013 Baylor College of Medicine, Houston, TX; 3Pediatric Cardiothoracic Surgery, Cincinnati Children’s Hospital Medical Center, Cincinnati, OH; 4Pharmacy, Texas Children’s Hospital, Houston, TX; 5Department of Pediatrics, Section of Infectious Disease, Texas Children’s Hospital, Baylor College of Medicine, Houston, TX. Purpose: Infectious complications constitute a major cause of morbidity and mortality in adult patients supported with left ventricular assist devices (VADs). Incidence and patient outcomes related to infectious complications in pediatric patients on VAD support remain largely unknown. The aim of this study was to determine the incidence of infection among pediatric VAD recipients and to characterize the microbiology, associated risk factors and clinical outcome. Methods and Materials: We conducted a retrospective chart review of all patients undergoing VAD support for Z2 weeks at Texas Children’s Hospital from June 1999-December 2011. Infections were categorized as VAD-specific, VAD-related or non-VAD related using the International Society for Heart and Lung Transplantation (ISHLT) definitions for VAD infections. Results: Fifty-two VADs were implanted in 51 patients. Thirty-five (69%) patients had 92 infections on VAD support and included: 10 VAD-specific infections, 23 VAD-related infections, and 59 non-VAD infections. Overall rate of VAD-infections (specificþrelated) was 9.4/ 1000 days of VAD support. The most common pathogens were Staphylococcus aureus, coagulase negative staphylococci, Pseudomonas aeruginosa and Candida species. Three (37.5%) of 8 deaths occurring during VAD support were directly related to infections. Prolonged hospital stay (p¼0.005) and prior cardiac transplantation with rejection (p¼0.018) were significantly associated with development of VAD-infections. Conclusions: Infectious complications are common in pediatric patients undergoing VAD support. By identifying and controlling risk factors, it may be possible to reduce the occurrence of VAD infections in pediatric patients.
212 Electronic medical records were reviewed for infection details to supplement data available in our transplant databases. DLIs were defined according to Hannan, et al (2011). Results: From 7/04 to 7/11, 181 LVADs were implanted and 44 DLIs occurred during 304 person-years of LVAD support. Hazard for DLI after LVAD was constant at 2.2%/mo but increased to 10%/mo from 6-9 months. Pre-, peri-, post-operative variables did not correlate with risk for DLI. ‘‘Trauma’’ to drive line was described in 20/44 (45%) of DLI vs. 3/137 (2.1%) without DLI but this was not routinely documented. DLI doubled the risk for death before transplant (HR 2.16, CI95% 1.18, 3.96). Pseudomonas aeruginosa (PA) was the most common pathogen and often replaced initial pathogens over time. Superficial DLI progressed at a mean 173d (61-362d) to deep DLI in 9/39 (23%), 8/9 due to PA, 1/9 Staphylococcus aureus. Multidrug resistance emerged in 4 PA and 3 other gram negative DLIs. Other complications of antibiotics rarely occurred. Conclusions: DLI prevalence accumulates over time in patients on LVAD support. Our data demonstrate a peak hazard of DLI from 6-9 months after implantation. DLI is an independent predictor of death before transplantation in our population. PA, more than other organisms, progressed to deep DLI and became more drug resistant over time on support. 211 Infectious Complications and Outcomes in Children Supported with Left Ventricular Assist Devices A.G. Cabrera,1 M.S. Khan,2 D.L.S. Morales,3 D.W. Chen,2 B.S. Moffett,4 J.F. Price,1 W.J. Dreyer,1 S.W. Denfield,1 A. Jeewa,1 C.D.Fraser Jr.,2 J.G. Vallejo.5 1Department of Pediatrics, Section of Pediatric Cardiology, Texas Children’s Hospital, Baylor College of Medicine, Houston, TX; 2Michael E. DeBakey Department of Surgery, Division of Congenital Heart Surgery, Texas Children’s Hospital,
Thrombotic Complications Increase after Percutaneous Site/Pocket Infection in Patients with Left Ventricular Assist Devices: An INTERMACS Analysis M.R. Shah,1 D.C. Naftel,2 M.A. Miller,1 J.T. Baldwin,1 E.D. Feller,3 S.S. Desai,4 R. John,5 R.L. Kormos,6 W.L. Holman.2 1National Heart, Lung, and Blood Institute, Bethesda, MD; 2University of Alabama at Birmingham, Birmingham, AL; 3University of Maryland Medical Center, Baltimore, MD; 4Inova Fairfax Hospital, Falls Church, VA; 5 University of Minnesota, Minneapolis, MN; 6University of Pittsburgh Medical Center, Pittsburgh, PA. Purpose: Previous studies suggest infection may confer risk for thrombotic complications (TC) after LVAD. We hypothesized that in discharged (disch) LVAD patients (pts), there is an increased rate of TC within 6 weeks (wks) after a major infection. Methods and Materials: We reviewed INTERMACS (4/2008 to 6/2012) to identify disch pts with continuous flow LVAD, with Z1 infection after disch. INTERMACS definitions were used. Infection was categorized as: 1) pump infection (pump); 2) percutaneous site/pocket
Percutaneous Site/Pocket Infection
Total TC Neurological dysfunction Pump thrombosis Non-CNS TC Hemolysis
Pre-infection (n; rate)
Post-infection (n; rate) p-value
12; 1.8 6; 0.9
33; 4.9 18; 2.7
0.005 0.01
2; 0.3
7; 1.0
0.06
0; 0.0 4; 0.6
1; 0.2 7; 1.04
– 0.48
Abstracts
S85
infection (device related); 3) sepsis; and, 4) localized non-device infection (non-device). TC were: 1) neurological dysfunction; 2) pump thrombosis; 3) arterial non-central nervous system and, 4) hemolysis (plasma free hemoglobin 440 mg/L). TC rates (TC per 100 mo of pt exposure) 6 wks pre-infection were compared to rates 6 wks postinfection. P-values were calculated with McNemar’s test. Results: Of 4124 disch LVAD pts, 1253 (30%) had a major infection post-disch (pump, n¼5; device related, n¼517; sepsis, n¼203; nondevice, n¼528). Key features of infected pts (mean⫾SD and %) were: age (54.6⫾8.3 yrs.), male (75.4%), diabetes (43.8%), destination therapy (27.9%). Last INR prior to TC was 2.1⫾1.3. There was no significant difference in TC rate after an infection vs pre-infection (6.63 vs 6.27; p¼0.67). When analyzed by category of infection, there was a higher rate of TC after device related infection. Conclusions: In disch LVAD pts there was a significantly higher rate of TC driven by neurological dysfunction and pump thrombosis within 6 wks of a device related infection. There was no significant association between other infections and TC. This data suggests increased risk of TC following device related infection. 213 Incidence and Characterization of Percutaneous Lead Damage in the Heartmate II Left Ventricular Assist Device D. Kalavrouziotis,1 A. Massiello,2 M.Z. Tong,1 R.C. Starling,3 R. Fryc,4 G. Heatley,4 D.J. Farrar,4 N. Moazami.1 1Department of Thoracic and Cardiovascular Surgery, Cleveland Clinic, Cleveland, OH; 2Department of Biomedical Engineering, Lerner Research Institute, Cleveland Clinic, Cleveland, OH; 3Department of Cardiovascular Medicine, Cleveland Clinic, Cleveland, OH; 4Thoratec Corporation, Pleasanton, CA. Purpose: The incidence of percutaneous lead failure due to wear and traumatic fracture among patients supported with a Heartmate (HM) II left ventricular assist device (LVAD) has not been previously reported. Methods and Materials: The study is a retrospective review of all HM II LVAD percutaneous driveline malfunctions reported to Thoratec Corporation. Categorization of wear and failure by location and severity was performed and the incidence of adverse clinical outcomes as well as the management of the dysfunctional leads were examined. Results: Between 2004 and October 2012, 12,969 HM II pumps were implanted worldwide. The incidence of percutaneous lead malfunction was 1,418 events occurring in 1,198 pumps (9.2%) over a cumulative support period of 13,932 patient-years (maximum ¼ 7.3 years). Reported lead damage was mostly in the externalized part of the cable (87.2%). Lead malfunction was managed with external reinforcement in 7.5% of the implanted pumps, while mortality or significant morbidity including pump exchange or urgent transplant, or more complex endcable replacement occurred in 2.3%. Percutaneous lead malfunction not associated with adverse events was managed by clamshell reinforcement of the connector strain relief or by tape/silicone cable reinforcement. The cumulative incidence of internal lead failures has decreased with two lead design revisions: 1.6% in 2004 (original design) vs. 0.01% in 2010 (redesigned internal pump end bend relief). Conclusions: The incidence of HM II percutaneous lead malfunction associated with serious adverse events is low and the majority of patients who experience lead damage have been successfully managed by external repairs. The incidence of both internal and external lead failures has diminished since 2004 with progressive iterations and improvements in lead design. 214 Predisposition to Infection and Bleeding in Mechanical Assist Devices (LVADs) Recepients after Dental Procedures: A Single Center Experience K.C. McCants, J.B. McCants, P.S. Combs, P. Raheja, M.S. Slaughter, E. Birks. Advanced Heart Failure and Transplant, University of Louisville & Jewish Hospital, Louisville, KY. Purpose: Currently more than 7,000 patients with heart failure are supported with LVADs. Few patients with LVADs undergo dental
procedures and are at risk of bleeding due to their anticoagulation/ antiplatelet medication profile and acquired VWF deficiency. Many centers follow the AHA guidelines for the prophylaxis of bacterial endocarditis in patients with prosthetic devices. Though these guidelines are comprehensive, they do not specifically address prophylactic recommendations for patients with LVADs. We would assume that these guidelines hold true for LVADs; however there is no current evidence. Methods and Materials: To assess the risk of infections and bleeding complications during dental procedures in patients with LVADs, we designed a protocol based pilot study of LVAD patients using the AHA guidelines. We followed 19 patients respective clinical course for administration of preprocedural antibiotics, continuation of their