Jonathan W. Haft

Extracorporeal Life Support for Severe Acute Respiratory Distress Syndrome in Adults

Objective:Severe acute respiratory distress syndrome (ARDS) is associated with a high level of mortality. Extracorporeal life support (ECLS) during severe ARDS maintains oxygen and carbon dioxide gas exchange while providing an optimal environment for recovery of pulmonary function. Since 1989, we have used a protocol-driven algorithm for treatment of severe ARDS, which includes the use of ECLS when standard therapy fails. The objective of this study was to evaluate our experience with ECLS in adult patients with severe ARDS with respect to mortality and morbidity. Methods:We reviewed our complete experience with ELCS in adults from January 1, 1989, through December 31, 2003. Severe ARDS was defined as acute onset pulmonary failure, with bilateral infiltrates on chest x-ray, and PaO2/fraction of inspired oxygen (FiO2) ratio ≤100 or A-aDO2 >600 mm Hg despite maximal ventilator settings. The indication for ECLS was acute severe ARDS unresponsive to optimal conventional treatment. The technique of ECLS included veno-venous or veno-arterial vascular access, lung “rest” at low FiO2 and inspiratory pressure, minimal anticoagulation, and optimization of systemic oxygen delivery. Results:During the study period, ECLS was used for 405 adult patients age 17 or older. Of these 405 patients, 255 were placed on ECLS for severe ARDS refractory to all other treatment. Sixty-seven percent were weaned off ECLS, and 52% survived to hospital discharge. Multivariate logistic regression analysis identified the following pre-ELCS variables as significant independent predictors of survival: (1) age (P = 0.01); (2) gender (P = 0.048); (3) pH ≤7.10 (P = 0.01); (4) PaO2/FiO2 ratio (P = 0.03); and (5) days of mechanical ventilation (P < 0.001). None of the patients who survived required permanent mechanical ventilation or supplemental oxygen therapy. Conclusion:Extracorporeal life support for severe ARDS in adults is a successful therapeutic option in those patients who do not respond to conventional mechanical ventilator strategies.

The Development of Aortic Insufficiency in Left Ventricular Assist Device-Supported PatientsClinical Perspective

Background—Aortic insufficiency (AI) following left ventricular assist device (LVAD) placement can affect device performance. The aim of this study was to examine AI development following LVAD implantation. Methods and Results—Echocardiograms (n=315) from 78 subjects undergoing HeartMate-XVE (n=25 [32%]) or HeartMate-II (n=53 [68%]) implantations from 2004 to 2008 were reviewed. Studies were obtained preoperatively and at 1, 3, 6, 12, 18, and 24 months after surgery. AI was graded on an interval scale (0=none, 0.5=trivial, 1=mild, 1.5=mild-moderate, 2=moderate, 2.5=moderate-severe, 3=severe), and the change in AI at follow-up was analyzed with significance tests. Kaplan–Meier estimates for freedom from moderate or worse AI at follow-up were generated. Mixed-model linear regression was used to identify correlates of AI progression during LVAD support. The median (25th, 75th percentile) duration of LVAD support was 239 (112, 455) days, and preoperative AI grade was 0.0 (0.0, 0.0). At 6 months, 89±4% of subjects (n=49 at risk) were free from moderate or worse AI, but this was reduced to 74±7% (n=29 at risk) and 49±13% (n=13 at risk) by 12 and 18 months, respectively. Correlates (slope±SE) of AI progression included female sex (0.002±0.001; P=0.01), smaller body surface area (−0.003±0.001 per m2; P=0.0017), and HeartMate-II model type (0.002±0.001; P=0.039). Correlates (&bgr;±SE) of progressive AI on postoperative echocardiogram included increasing aortic sinus diameter (0.04±0.01 per mm; P=0.001), an aortic valve that remained closed (0.42±0.06; P<0.001) or only intermittently opened (0.34±0.09; P<0.001), and lower left ventricular diastolic (−0.002±0.0004 per cm3; P<0.001) and systolic (−0.002±0.0004 per cm3; P<0.001) volumes. Conclusions—AI progresses over time in LVAD-supported patients. As we move toward an era of long-term cardiac support, more studies are needed to determine the clinical significance of these findings.

The Development of Aortic Insufficiency in Left Ventricular Assist Device-Supported Patients

Background—Aortic insufficiency (AI) following left ventricular assist device (LVAD) placement can affect device performance. The aim of this study was to examine AI development following LVAD implantation. Methods and Results—Echocardiograms (n=315) from 78 subjects undergoing HeartMate-XVE (n=25 [32%]) or HeartMate-II (n=53 [68%]) implantations from 2004 to 2008 were reviewed. Studies were obtained preoperatively and at 1, 3, 6, 12, 18, and 24 months after surgery. AI was graded on an interval scale (0=none, 0.5=trivial, 1=mild, 1.5=mild-moderate, 2=moderate, 2.5=moderate-severe, 3=severe), and the change in AI at follow-up was analyzed with significance tests. Kaplan–Meier estimates for freedom from moderate or worse AI at follow-up were generated. Mixed-model linear regression was used to identify correlates of AI progression during LVAD support. The median (25th, 75th percentile) duration of LVAD support was 239 (112, 455) days, and preoperative AI grade was 0.0 (0.0, 0.0). At 6 months, 89±4% of subjects (n=49 at risk) were free from moderate or worse AI, but this was reduced to 74±7% (n=29 at risk) and 49±13% (n=13 at risk) by 12 and 18 months, respectively. Correlates (slope±SE) of AI progression included female sex (0.002±0.001; P=0.01), smaller body surface area (−0.003±0.001 per m2; P=0.0017), and HeartMate-II model type (0.002±0.001; P=0.039). Correlates (&bgr;±SE) of progressive AI on postoperative echocardiogram included increasing aortic sinus diameter (0.04±0.01 per mm; P=0.001), an aortic valve that remained closed (0.42±0.06; P<0.001) or only intermittently opened (0.34±0.09; P<0.001), and lower left ventricular diastolic (−0.002±0.0004 per cm3; P<0.001) and systolic (−0.002±0.0004 per cm3; P<0.001) volumes. Conclusions—AI progresses over time in LVAD-supported patients. As we move toward an era of long-term cardiac support, more studies are needed to determine the clinical significance of these findings.

Model for End-Stage Liver Disease Score Predicts Left Ventricular Assist Device Operative Transfusion Requirements, Morbidity, and Mortality

Background— The Model for End-Stage Liver Disease (MELD) predicts events in cirrhotic subjects undergoing major surgery and may offer similar prognostication in left ventricular assist device candidates with comparable degrees of multisystem dysfunction. Methods and Results— Preoperative MELD scores were calculated for subjects enrolled in the University of Michigan Health System (UMHS) mechanical circulatory support database. Univariate and multiple regression analyses were performed to investigate the ability of patient characteristics, laboratory data (including MELD scores), and hemodynamic measurements to predict total perioperative blood product exposure and operative mortality. The ability of preoperative MELD scores to predict operative mortality was evaluated in subjects enrolled in the Interagency Registry of Mechanically Assisted Circulatory Support (INTERMACS), and results were compared with those from the UMHS cohort. The mean±SD MELD scores for the UMHS (n=211) and INTERMACS (n=324) cohorts were 13.7±6.1 and 15.2±5.8, respectively, with 29 (14%) and 19 (6%) perioperative deaths. In the UMHS cohort, median total perioperative blood product exposure was 74 units (25th and 75th percentiles, 44 and 120 units). Each 5-unit MELD score increase was associated with 15.1±3.8 units (&bgr;±SE) of total perioperative blood product exposure. Each 10-unit increase in total perioperative blood product exposure increased the odds of operative death (odds ratio, 1.05; 95% confidence interval, 1.01 to 1.10). Odds ratios, measuring the ability of MELD scores to predict perioperative mortality, were 1.5 (95% confidence interval, 1.1 to 2.0) and 1.5 (95% confidence interval, 1.1 to 2.1) per 5 MELD units for the UMHS and INTERMACS cohorts, respectively. When MELD scores were dichotomized as ≥17 and <17, risk-adjusted Cox proportional-hazard ratios for 6-month mortality were 2.5 (95% confidence interval, 1.2 to 5.3) and 2.5 (95% confidence interval, 1.1 to 5.4) for the UMHS and INTERMACS cohorts, respectively. Conclusion— The MELD score identified left ventricular assist device candidates at high risk for perioperative bleeding and mortality.

Hemodynamic and exercise performance with pulsatile and continuous-flow left ventricular assist devices.

Background— Continuous-flow rotary pumps with axial design are increasingly used for left ventricular assist support. The efficacy of this design compared with pulsatile, volume displacement pumps, with respect to characteristics of left ventricular unloading, and exercise performance remains largely unstudied. Methods and Results— Thirty-four patients undergoing implantation with a pulsatile, volume displacement pump operating in a full-to-empty cycle (HeartMate XVE; Thoratec Inc, Pleasanton, Calif; n=16) or continuous-flow rotary pump with an axial design operating at a fixed rotor speed (HeartMate II; Thoratec Inc; n=18) were evaluated with right heart catheterization and echocardiography preoperatively and at 3 months postoperatively and cardiopulmonary exercise testing 3 months postoperatively. Support with either the XVE or II resulted in significant (P<0.05) increases in cardiac output and reduction in mean pulmonary artery and pulmonary wedge pressures. Exercise capacity at 3 months was similar between groups (% predicted peak o2−XVE: 46.8±10.2 versus II: 49.1±13.6). Echocardiography at 3 months demonstrated a significantly (P<0.05) greater reduction in left ventricular end-diastolic volume (−49±16% versus −35±20%), left ventricular end-systolic volume (−59±20 versus −37±21%), and percent mitral valve regurgitant volume (−99±2% versus −52±56%) for the XVE compared with II, respectively. Conclusions— The HeartMate XVE or II provided equivalent degrees of hemodynamic support and exercise capacity. The XVE was associated with greater left ventricular volume unloading. Characteristics of left ventricular pressure and volume unloading between these pump designs and mode of operation do not influence early exercise performance.

Hemolysis: A harbinger of adverse outcome after left ventricular assist device implant

BACKGROUND The clinical relevance of elevated serum markers of hemolysis during left ventricular assist device (LVAD) support has not been fully ascertained. METHODS Lactate dehydrogenase (LDH) and serum free hemoglobin (sfHg) values were tallied monthly in 182 patients on HeartMate II (Thoratec, Pleasanton, CA) LVAD support. Peak values for each marker were identified, and 2 hemolysis definitions were applied to the cohort: Hemolysis according to Interagency Registry for Mechanically Assisted Circulatory Support (INTERMACS) criteria (sfHg > 40 mg/dl with signs/symptoms) and/or hemolysis defined by an LDH ≥ 600 IU/liter (2.5-times the upper limit of laboratory normal). Kaplan-Meier survival free from death, urgent United Network of Organ Sharing 1A transplant for thrombosis, device exchange for thrombosis, and stroke/peripheral embolism was estimated, and Cox hazard ratios (HR) with the 95% confidence interval (95% CI) were calculated. Areas under the receiver-operating characteristic curves (AUCs) for predicting 1-year event-free survival were calculated. RESULTS Hemolysis occurred in 32 patients (18%) by INTERMACS criteria and in 68 (37%) patients by LDH criteria. Over a median (25(th), 75(th)) support of 427 days (245, 793 days), there were 78 events. One year event-free survival after the onset of INTERMACS-defined hemolysis was 16% ± 8.3% compared with 85% ± 3.2% in non-hemolyzers (HR, 14.7; 95% CI, 7.9-27; AUC 0.70 ± 0.05; p < 0.001; ). One year event-free survival after the onset of LDH-defined hemolysis was 32% ± 7.2% compared with 89% ± 3.2% in those with persistent LDH values < 600 IU/liter (HR, 8.0; 95% CI, 4.4-14; AUC 0.87 ± 0.04; p < 0.001). Patients who met the LDH hemolysis definition had longer times from hemolysis onset to clinical events and larger magnitudes of risk for embolism and device exchange for thrombosis than those with INTERMACS hemolysis. CONCLUSIONS Serum hemolysis marker elevations are associated with increased events in LVAD patients. LDH monitoring provides an earlier diagnosis of adverse events than sfHg, supporting need for a new INTERMACS definition of VAD-associated hemolysis.

Extracorporeal life support: experience with 2,000 patients.

This is a review of the University of Michigan experience with extracorporeal life support (ECLS) also known as extracorporeal membrane oxygenation (ECMO). Two thousand patients were managed with ECMO from 1973 to 2010. The first 1,000 patients were reported previously. Of the 2,000 patients, 74% were weaned from ECLS, and 64% survived to hospital discharge. In patients with respiratory failure, survival to hospital discharge was 84% in 799 neonates, 76% in 239 children, and 50% in 353 adults. Survival in patients with cardiac failure was 45% in 361 children and 38% in 119 adults. ECLS during extracorporeal cardiopulmonary resuscitation was performed in 129 patients, with 41% surviving to discharge. Survival decreased from 74 to 55% between the first and second 1,000 patients. The most common complication was bleeding at sites other than the head, with an incidence of 39%, and the least frequent complication was pump malfunction, with a 2% incidence. Intracranial bleeding or infarction occurred in 8% of patients, with a 43% survival rate. This is the largest series of ECLS at one institution reported in the world to date. Our experience has shown that ECLS saves lives of moribund patients with acute pulmonary and cardiac failure in all age groups.

Two Decades' Experience With Interfacility Transport on Extracorporeal Membrane Oxygenation

BACKGROUND Interfacility transport of patients on extracorporeal membrane oxygenation (ECMO) has been performed in large numbers at only a few programs. Limited data are available on outcomes after ECMO transport to justify expanding or discontinuing these programs. METHODS This was a retrospective review of a 20-year, single-institution experience with interhospital ECMO transport as well as a systematic review of reports of transfers of patients on ECMO. Results of both were compared with historical data from the international registry of the Extracorporeal Life Support Organization (ELSO). RESULTS Between 1990 and 2012, ECMO was used to facilitate transport of 221 patients to our institution, and 135 (62%) survived to discharge. Review of an additional 27 case series describing ECMO transport of 643 patients showed an overall survival of 61%. After stratifying by age and primary indication for ECMO, survival of transported patients was not significantly different compared with all ECMO patients in the ELSO registry, with the exception of pediatric patients treated for respiratory failure (transported patients in this category had higher survival than those in the ELSO registry). CONCLUSIONS Interfacility transport on ECMO is feasible and can be accomplished safely in the critically ill. Survival of transported patients is comparable to age-matched and treatment-matched ECMO patients at large.

Consequences of aortic insufficiency during long-term axial continuous-flow left ventricular assist device support

BACKGROUND Although left ventricular assist device (LVAD) management strategies are undertaken to reduce the development of aortic insufficiency (AI), the effect of AI on patient morbidity and mortality is not known. METHODS Patients undergoing HeartMate II (Thoratec, Pleasanton, CA) implant were prospectively monitored with serial echocardiograms. Kaplan-Meier methods and log-rank testing were used to estimate and compare mortality and freedom from moderate or worser right ventricular hypokinesis (RVHK), moderate or worse mitral regurgitation (MR), and hemolysis according to AI severity. Mixed modelling was used to examine for correlates of AI development in the pre-operative and post-operative setting and to investigate the effect of AI on post-operative MR and RVHK. RESULTS There were 930 echocardiograms completed in 166 patients. During 291 person-years of follow-up, mild-moderate or worse AI developed in 70 (0.38 persons per year [PPY]), moderate or worse AI in 36 (0.17 PPY), moderate-severe AI in 11 (0.039 PPY), and severe AI in 2 (0.0069 PPY). Overall 2-year survival and 2-year survival after onset of moderate or worse AI was 87% ± 6.2% and 65% ± 11%, respectively, compared with 76% ± 5.1% and 76% ± 5.1%, respectively, in those with less AI (p = 0.57). Patients with moderate AI were not more likely to develop MR, hemolysis events, or worsening RVHK, but patients with pre-existing RVHK appeared to be less tolerant of AI. Three of 35 deaths were directly attributed to AI. No reoperations were performed solely for AI. CONCLUSIONS AI is common after LVAD implant but did not affect survival in this cohort. Except in those with significant RV dysfunction, this calls into question need for echocardiogram-guided device settings to ensure aortic valve opening.

Use of venovenous extracorporeal membrane oxygenation and an atrial septostomy for pulmonary and right ventricular failure.

BACKGROUND Right ventricular failure is a major contributor to morbidity and mortality on the lung transplant waiting list. This study was designed to evaluate the effectiveness of an atrial septostomy with venovenous extracorporeal membrane oxygenation (VV-ECMO) as a novel potential bridge to transplantation. METHODS Adult sheep (58±3 kg; n=12) underwent a clamshell thoracotomy and instrumentation to measure all relevant pressures and cardiac output (CO). Sheep with tricuspid insufficiency (TI [n=5]) and without tricuspid insufficiency (ØTI [n=7]) were examined. After creation of a 1-cm atrial septal defect and initiating VV-ECMO, the pulmonary artery (PA) was banded to allow progressive reduction of pulmonary blood flow, and data were collected. RESULTS The CO in both groups remained unchanged from baseline at all pulmonary blood flow conditions. With TI, the CO was 5.1±1.2 L/min at baseline versus 5.1±1.2 L/min with a fully occluded PA (p=0.99). For ØTI, the CO was 4.5±1.4 L/min at baseline versus 4.5±1.2 L/min with no pulmonary blood flow (p=0.99). Furthermore, CO was not affected by the presence of TI (p=0.76). Mean right ventricular pressures were significantly lower in the TI group (TI=20.2±11 mm Hg versus ØTI=29.9±8.9 mm Hg; p<0.00001). Right and left atrial mean arterial pressures were not different between both groups (p>0.5). Lastly, VV-ECMO maintained normal blood gases, with mean O2 saturations of 99% ± 4.1% in both groups. CONCLUSIONS Right to left atrial shunting of oxygenated blood with VV-ECMO is capable of maintaining normal systemic hemodynamics and normal arterial blood gases during high right ventricular afterload dysfunction.