Inhaled nitric oxide after left ventricular assist device implantation: A prospective, randomized, double-blind, multicenter, placebo-controlled trial Evgenij Potapov, MD, PhD,a Dan Meyer, MD,b Madhav Swaminathan, MD,c Michael Ramsay, MD,d Aly El Banayosy, MD,e Christoph Diehl, MD,f Bryan Veynovich, DO,g Igor D. Gregoric, MD,h Marian Kukucka, MD,a Tom W. Gromann, MD,a Nandor Marczin, MD, PhD,i Kanti Chittuluru, MD,j James S. Baldassarre, MD,j Mark J. Zucker, MD,k and Roland Hetzer, MD, PhDa From aDeutsches Herzzentrum Berlin, Berlin, Germany; bUniversity of Texas Southwestern/St Paul Medical Center, Dallas, Texas; cDuke University Medical Center, Durham, North Carolina; dBaylor University Medical Center, Dallas, Texas; ePenn State Milton S. Hershey Medical Center Heart and Vascular Institute, Hershey, Pennsylvania; fHerz- und Diabeteszentrum Nordrhein-Westfalen, Bad Oeynhausen, Germany; gAllegheny General Hospital, Pittsburgh, Pennsylvania; hTexas Heart Institute, Houston, Texas; iImperial College London, Royal Brompton and Harefield NHS Foundation Trust, London, United Kingdom; jDepartment of Clinical Development, Ikaria, Clinton, New Jersey; and kNewark Beth Israel Medical Center, Newark, New Jersey.
KEYWORDS: heart-assist device; heart failure; nitric oxide; ventilation
BACKGROUND: Used frequently for right ventricular dysfunction (RVD), the clinical benefit of inhaled nitric oxide (iNO) is still unclear. We conducted a randomized, double-blind, controlled trial to determine the effect of iNO on post-operative outcomes in the setting of left ventricular assist device (LVAD) placement. METHODS: Included were 150 patients undergoing LVAD placement with pulmonary vascular resistance ⱖ200 dyne/sec/cm–5. Patients received iNO (40 ppm) or placebo (an equivalent concentration of nitrogen) until 48 hours after separation from cardiopulmonary bypass, extubation, or upon meeting study-defined RVD. For ethical reasons, crossover to open-label iNO was allowed during the 48-hour treatment period if RVD criteria were met. RESULTS: RVD criteria were met by 7 of 73 patients (9.6%; 95% confidence interval, 2.8 –16.3) in the iNO group compared with 12 of 77 (15.6%; 95% confidence interval, 7.5–23.7) who received placebo (p ⫽ 0.330). Time on mechanical ventilation decreased in the iNO group (median days, 2.0 vs 3.0; p ⫽ 0.077), and fewer patients in the iNO group required an RVAD (5.6% vs 10%; p ⫽ 0.468); however, these trends did not meet statistical boundaries of significance. Hospital stay, intensive care unit stay, and 28-day mortality rates were similar between groups, as were adverse events. Thirty-five patients crossed over to open-label iNO (iNO, n ⫽ 15; placebo, n ⫽ 20). Eighteen patients (iNO, n ⫽ 9; placebo, n ⫽ 9) crossed over before RVD criteria were met. CONCLUSIONS: Use of iNO at 40 ppm in the perioperative phase of LVAD implantation did not achieve significance for the primary end point of reduction in RVD. Similarly, secondary end points of time on mechanical ventilation, hospital or intensive care unit stay, and the need for RVAD support after LVAD placement were not significantly improved. J Heart Lung Transplant 2011;30:870 – 8 © 2011 International Society for Heart and Lung Transplantation. All rights reserved.
Reprint requests: Evgenij Potapov, MD, PhD, Klinik für Herz-, Thorax- und Gefässchirurgie, Augustenbuger Platz 1, 13353 Berlin, Germany. Telephone: ⫹49-30-4593-2065. Fax: ⫹49-30-4593-2079. E-mail address: [email protected]
CLINICAL TRIALS IDENTIFIER: NCT00060840; www.clinicaltrials.gov/ct2/results?term⫽NCT00060840
1053-2498/$ -see front matter © 2011 International Society for Heart and Lung Transplantation. All rights reserved. doi:10.1016/j.healun.2011.03.005
Potapov et al.
Use of Inhaled Nitric Oxide After LVAD Placement
In patients with severe congestive heart failure, placement of a left ventricular assist device (LVAD) is known to normalize hemodynamics, improve end-organ dysfunction, and increase exercise tolerance.1 At present, long-term LVAD support is approved as a treatment for patients with end-stage heart failure as a bridge to transplantation or as a permanent support, otherwise known as destination therapy. For many reasons, isolated LVAD support is preferred to biventricular support or total artificial heart placement in patients with end-stage cardiac failure.1,2 Unfortunately, LVAD placement is frequently complicated by right ventricular dysfunction (RVD),3 which has been reported in 5% to 39% of patients.2,4 –7 When manifest, RVD may result in circulatory failure, delayed weaning from cardiopulmonary bypass (CPB), compromised LVAD flow, and poor tissue perfusion, all associated with significant increases in overall morbidity and mortality.4,5,8 Although preoperative risk estimation is improving,5,6,9 there is no clear agreement on the best markers of risk.5 Tricuspid valve and right ventricular (RV) function/geometry,10,11 grade of preoperative cardiogenic shock, preoperative management, patient selection, and even sex seem to play a role in assigning risk.4,5 Numerous small studies have investigated the effect of inhaled nitric oxide (iNO) on hemodynamics in various cardiac surgeries, demonstrating that iNO can reduce pulmonary vascular resistance (PVR), thereby reducing the likelihood of high PVR-associated RVD.12–21 Numerous institutions have used iNO for patients at risk for RVD post-operatively.4,22 No large, prospective, randomized, controlled trials have assessed the utility of the prophylactic use of iNO in preventing post-operative RVD in patients undergoing LVAD placement. The purpose of this prospective, randomized, controlled trial was to study the safety and efficacy of iNO in preventing RVD clinical outcomes after LVAD implantation. The primary study end point was the incidence of RVD, based on study-defined criteria, within 48 hours of surgery.
Methods This study was conducted at 8 centers in the United States and Germany between September 2003 and March 2008. Each country’s national health authority and local Investigational Review Board/Ethics Committees approved the study. Consent of participating patients was obtained according to local requirements. The study was designed and monitored by a Steering Committee (Appendix 1). An independent Data and Safety Monitoring Committee reviewed unblinded safety data. Investigators are listed in Appendix 2. Patients aged ⬎18 years scheduled to undergo their first LVAD placement were screened for enrollment and were eligible if they had a PVR of ⱖ200 dyne/sec/cm–5 before LVAD placement. Exclusion criteria included patients scheduled for biventricular assist device (BiVAD) surgery; patients being supported with a temporary BiVAD or extracorporeal membrane oxygenation, if LVAD placement was to be performed without CPB; patients with congestive heart failure due to giant cell myocarditis or restrictive
cardiomyopathy; and patients receiving investigational drugs that were expected to change systemic vascular resistance or PVR. Patients received iNO (40 ppm) or placebo (an equivalent nitrogen concentration), initiated at least 5 minutes before the first weaning attempt from CPB and continued until the patient (1) was extubated, (2) reached a study end point, or (3) was treated for 48 hours, whichever came first. For ethical reasons, patients had the option to cross over to open-label iNO immediately and remain on iNO for up to 14 days, if they (1) failed to wean at least once from CPB due to hemodynamic failure, (2) still required pulmonary vasodilator support at 48 hours, or (3) met predefined RVD criteria (Table 1). Nitrogen dioxide and methemoglobin (MetHgb) levels were monitored throughout treatment. The quantity of blood products used was recorded from the start of the study drug until Day 28. All adverse events (AEs) were documented during the study drug and open-label iNO administration. Serious AEs were documented during the entire study period. Patients were randomized to iNO (40 ppm) or placebo, in a block size of 4, stratified by site and LVAD type (axial or pulsatile). Study drug cylinders were pre-randomized and labeled to identify the randomized patient number without revealing the treatment group. Blinded INOvent delivery systems (Ikaria, Clinton, NJ) were used and were blinded both by a physical faceplate and electronic display modifications. Patients were weaned from mechanical ventilation and extubated according to standard medical care and hospital-specific protocol. No specific weaning criteria were defined for the study, but were institutionally based and conducted at the discretion of the participating centers. The primary efficacy end point was the number of patients who met the predefined RVD criteria (Table 1). Secondary efficacy end points were (1) days on mechanical ventilation, (2) number of intensive care unit days, (3) number of total hospital days, (4) number of patients requiring renal replacement therapy, (5) quan-
Table 1 Criteria
Study-Defined Right Ventricular Dysfunction
RVD criteria were defined as any of the following: ● Death ● Inability to wean from CPB, or ● Any 2 of the following (sustained for 15 minutes after complete withdrawal from CPBa): X LVFRI ⱕ2.0 liters/min/m2, calculated as LVAD flow divided by body surface area X Administration of ⱖ20 inotropic equivalents (IE): 10 g/kg/min dopamine, dobutamine, enoximone, or amrinone is equivalent to 10 IE 0.1 g/kg/min epinephrine or norepinephrine is equivalent to 10 IE 1 g/kg/min milrinone is equivalent to 15 IE 0.1 U/min vasopressin is equivalent to 10 IE X MAP ⱕ55 mm Hg X CVP ⱖ16 mm Hg X SvO2 ⱕ55% CPB, cardiopulmonary bypass; CVP, central venous pressure; IE, inotropic equivalents; LVAD, left ventricular assist device; LVFRI, left ventricular flow rate (pump) index; MAP, mean arterial pressure; RVD, right ventricular dysfunction; SvO2, mixed venous oxygen saturation. a Reinitiation of CPB to correct bleeding or other technical issues was not considered failure to wean.
The Journal of Heart and Lung Transplantation, Vol 30, No 8, August 2011
tity of blood products used, (6) number of patients requiring right ventricular assist device (RVAD) by Day 28, and (7) survival at Day 28. Safety end points included changes in MetHgb levels and the incidence of AEs and serious AEs. Efficacy analyses were performed on the intent-to-treat (ITT) population (ie, all randomized patients). Safety analyses were performed on the safety population (ie, all patients who received study therapy). Additional analyses were performed on a post hoc, “per-protocol” population that included only those patients who met all entry criteria and received the study treatment. The primary efficacy end point was tested by Fisher’s exact test. Secondary efficacy variables with binary responses were analyzed using logistic regression. Other secondary efficacy variables were analyzed using an analysis of variance model with study site and treatment group as the main effects. A hierarchical gatekeeping strategy was planned for multiplicity adjustment in evaluating secondary efficacy end points. Estimated RVD rates of 50% in the placebo group and 25% and iNO group were used to initially assess a sample size of 55 per treatment arm (combined event rate, 37.5%). The total sample size was increased to 150, with 75 in each arm, based on the blinded event rate, which was found to be much lower (14.6%) over the course of the trial.
Results The study included 150 patients, with 73 randomized to the iNO group and 77 to the placebo groups. The study therapy was not administered to 13 patients (4 iNO, 9 placebo), resulting in 137 patients for the safety and per-protocol populations, comprising 69 in the iNO group and 68 in the placebo group (Figure 1). In the ITT population, the iNO group had a higher mean age compared with placebo; both groups were otherwise well matched in demographics and baseline characteristics (Table 2), including type of LVAD and baseline medications (Table 3). No patients were taking sildenafil or any other medications that would have affected outcomes in this study. Crossover to open-label iNO occurred in 15 of 73 patients (21%) in the iNO group and 20 of 77 (26%) in the placebo group. Of these 35 patients, 20 crossed over after meeting failure criteria or completing the protocol-required blinded study drug period. Another 18 patients crossed over before the study-defined crossover criteria were met; of these, 15 (6 iNO group and 9 placebo group patients, all from the same site) were crossed over by the physician for safety reasons (Figure 1). Within 8 hours, 9 of 15 iNO patients (60%) crossed over to open-label treatment compared with 18 of 20 placebo patients (90%). All placebogroup patients who crossed over did so within the first 28 hours, whereas those in the iNO group did so within 52 hours (Figure 2). Mean duration of blinded therapy was 19.0 ⫾13.0 and 16.7 ⫾ 13.2 hours for the iNO and placebo groups, respectively. Mean duration of open-label therapy was 65.6 ⫾ 105.9 hours. In this study, 89 of the 137 patients (65%) received iNO as blinded or open-label treatment.
The RVD criteria were met by 19 of 150 patients (12.7%), and a breakdown of criteria components is described in Table 4. As reported in Table 5, 7 of 73 patients in the iNO group (9.6%; 95% confidence interval, 2.8%– 16.3%) met RVD criteria compared with 12 of 77 patients receiving placebo (15.6%; 95% CI, 7.5%–23.7%); this difference was not statistically significant (p ⫽ 0.330). In the per-protocol analysis, 7 of 69 iNO patients (10.1%) and 12 of 68 placebo patients (17.6%) met RVD criteria within 48 hours (p ⫽ 0.226). Patients with a PVR index (PVRI) ⬍ 270 had very low rates of protocol-defined RVD compared with those with a higher PVRI. Because of the small number of patients receiving pulsatile devices, data stratified by LVAD type were not analyzed. A logistic regression of patients meeting RVD criteria in 48 hours was used to determine all significant, non-treatment covariates, including use of blood products (p ⫽ 0.856), hypertension (p ⫽ 0.145), diagnosed myocardial infarction (p ⫽ 0.536), and gender (p ⫽ 0.122); none of these covariates were significant. Table 5 also details the secondary outcome measures. Mean duration of mechanical ventilation was 5.37 ⫾ 7.72 in the iNO group and 11.10 ⫾ 24.81 days in the placebo group. Median number of days on mechanical ventilation was lower in the iNO group but did not reach statistical significance (2 vs 3 days; p ⫽ 0.077). Per-protocol analysis gave similar results: in the iNO vs placebo groups, the mean length of days on mechanical ventilation was 5.43 ⫾ 7.76 vs 11.09 ⫾ 25.00, respectively, and a median of 2 vs 3 days, respectively (p ⫽ 0.131). Findings were consistent across centers and sex. This finding did not correlate with length of stay in the ICU or in the hospital. RVAD implantation was required in 4 iNO patients (5.6%) and 7 placebo patients (10.0%) by Day 28 (p ⫽ 0.468). The amount of blood products used and the need for reoperation due to bleeding were similar in both groups. Both groups had comparable rates of AEs and serious AEs (Table 6). The AEs during the blinded portion of the study were also comparable. During double-blind treatment, 18 of the 137 patients (13.1%) experienced serious AEs: 7 of 69 iNO patients (10.1%) and 11 of 68 placebo patients (16.2%). No serious AEs in the iNO arm occurred in ⬎5% of patients, and overall event rates in various AE subgroups showed no obvious differences between the two groups. The most frequently reported serious AEs in iNO vs placebo groups were RV failure, occurring in 3 of 69 (4.3%) vs 3 of 68 (4.4%); postprocedural hemorrhage, 2 of 69 (2.9%) vs 3 of 68 (4.4%); and hemorrhage, 1 of 69 (1.4%) vs 3 of 68 (4.4%). In general, MetHgb levels were not higher in the iNO group vs the placebo group. Elevations in NO2 levels were also very rare.
Discussion Right heart dysfunction is still a major complication of LVAD placement.5,7 Elevated PVR is a pre-operative risk factor for morbidities,12,14 and the absence of pulmonary
Potapov et al.
Use of Inhaled Nitric Oxide After LVAD Placement
Patient disposition. iNO, inhaled nitric oxide; RV, right ventricular.
vasoreactivity confers a worse prognosis.23 Inhaled NO selectively reduces PVR in patients undergoing cardiac surgical procedures.12 Reduction in PVR decreases RV afterload, which in turn may prevent RVD. Numerous strategies have been used to avoid RVD after LVAD placement, including optimizing device placement timing by echocardiographic/clinical criteria,6,24 optimizing preoperative management25 with single or multiple inotropes
or pulmonary vasodilators, and by routinely using iNO after LVAD placement. Previous studies have shown that iNO use is beneficial for RVD management during LVAD placement. Argenziano et al12 studied patients with heart failure after CPB who required LVAD insertion and who had pulmonary hypertension with limited LVAD flow. Inhaled NO decreased mean pulmonary artery pressure (PAP) and in-
874 Table 2
The Journal of Heart and Lung Transplantation, Vol 30, No 8, August 2011 Baseline Characteristics of the Intent-to-Treat Population iNO (n ⫽ 73)
Age, mean (SD) years Gender, No. (%) Male Race, No. (%) White Black Otherb Underlying diagnosis, No. (%) Ischemic cardiomyopathy Hypertension Valvular heart disease Other LVAD type, No. (%) Axial Pulsatile IABP, No. (%) Organ function, mean (SD)a Serum creatinine, mg/dl Blood urea nitrogen, mg/dl AST, U/liter Total bilirubin, mg/dl Baseline hemodynamics, mean (SD) Cardiac index, liters/min/m2 PVR, dyne/sec/cm–5 CVP, mm Hg PCWP, mm Hg SvO2, %
Placebo (n ⫽ 77)
60 (82.2) 11 (15.1) 2 (2.7)
57 (74.0) 13 (16.9) 7 (9.1)
0.244 0.826 NA
18 21 19 48
19 29 19 56
(24.7) (37.7) (24.7) (72.7)
⬎0.99 0.299 0.854 0.380
37 (48.1) 36 (46.8) 12 (15.6)
0.507 NA 0.525
(24.7) (28.8) (26.0) (65.8)
41 (56.2) 31 (42.5) 15 (20.5) 1.5 50.7 200.7 1.9
(0.6) (37.2) (744.9) (2.7)
1.6 50.1 129.3 1.9
(0.7) (32.1) (375.5) (3.2)
0.604 0.923 0.465 0.937
2.2 303.7 12.7 23.0 69.6
(1.79) (17.41) (5.96) (8.31) (12.32)
1.8 337.8 14.4 26.7 65.5
(0.69) (25.75) (7.23) (9.19) (18.47)
0.343 0.660 0.134 0.013 0.183
AST, aspartate aminotransferase; CVP, central venous pressure; IABP, intra-aortic balloon pump; iNO, inhaled nitric oxide; LVAD, left ventricular assist device; NA, not applicable; PCWP, pulmonary capillary wedge pressure; PVR, pulmonary vascular resistance; SvO2, mixed venous oxygen saturation. a Proc T test used to calculate p-value. b Includes American Indian/Alaskan Native and Hispanic patients.
creased LVAD flow. Wagner et al14 investigated the effect of up to 40 ppm iNO in patients with RVD after LVAD insertion, and iNO significantly decreased PAP and PVR while increasing the cardiac index. Although small clinical studies have suggested the benefit of iNO in this setting, none have been adequate to provide good estimates of the magnitude of treatment effect. We initiated this large controlled study to assess iNO use in the peri-operative management of RVD after LVAD placement. Because evidence of iNO benefit has been shown and this therapy is a standard of care in most institutions, we had to allow patients to cross over to open-label iNO to prevent study drug–related mortality if RVD criteria were met. This may explain why the mortality rate was similar in both groups. At the request of the treating physician, 15 (6 iNO, 9 placebo) were crossed over without meeting protocoldefined crossover criteria. These patients may have been more likely to develop RVD later. Stratification by crossing over vs not crossing over was not a defined end point for the study. Figure 1 provides the number of patients who crossed over and who met failure criteria in 48 hours; outside of these data, no relevant analyses by this stratification were calculated.
Previous studies have reported RVD rates after LVAD of 10% to 40%.2,4 – 6 Our study had a lower 12.7% incidence of RVD, which was likely due to earlier intervention, aggressive pre-operative management, stringent study protocol follow-up, and to specific definitions of RVD used. No international agreement exists regarding the definition of RV failure. One of the most useful definitions of right heart failure in the LVAD literature is a combination of the need for RVAD and prolonged use of inotropes (usually ⬎14 days). This definition was not appropriate in this study, however, because of the retrospective nature of this definition (prolonged use of inotropes).4,5 To use this definition, the diagnosis of “RV failure,” established in most cases by Day 14 only, would preclude immediate cross over to open-label iNO in the case of RVD, thereby harming the patient. The RVD definition in our study was partly derived from prospectively measured hemodynamic criteria. For safety reasons, the period when 2 hemodynamic criteria may sustain and lead to diagnosis of RVD was considered to be 15 minutes; afterward, cross over was necessary. Contrary to previous definitions, this approach allowed investigators to respond and alter treatment quickly. Although death and
Potapov et al.
Use of Inhaled Nitric Oxide After LVAD Placement
Table 3 Baseline Medications in the Intent-to-Treat Population
Medication Antifibrinolytics Aminocaproic acid Aprotinin Tranexamic acid Inotropes and vasopressors Adrenergics and dopaminergics Dobutamine Others Phosphodiesterase inhibitors Milrinone Others Vasopressin and analogs Vasopressin Vasodilators Glyceryl trinitrate Others
iNO (n ⫽ 73) No. (%)
Placebo (n ⫽ 77) No. (%)
45 5 20 20 67
44 10 18 16 68
(61) (11) (44) (44) (91)
(57) (23) (41) (36) (88)
38 (57) 27 (71) 11 (29)
52 (76) 35 (67) 17 (33)
27 (40) 25 (93) 2 (11)
16 (24) 16 (100) 0 (0.0)
2 28 24 4
0 24 21 3
(3) (38) (86) (14)
(0) (31) (88) (13)
Failure Criteria Met in the Safety Population
Failure criteriaa p-Value 0.62
Death Failure to wean from CPB ⱖ 2 times LVFRI ⱕ 2.0 liters/min/m2 Administration of ⱖ 20 IE CVP ⱖ 16 mm Hg MAP ⱕ55 mm Hg SvO2 ⱕ55%
iNO Placebo Total (n ⫽ 69) (n ⫽ 68) (N ⫽ 137) No. (%) No. (%) No. (%) 0 (0)
3 0 3 2 3 1
6 0 5 3 1 3
9 0 8 5 4 4
(4.3) (0) (4.3) (2.9) (4.3) (1.4)
(8.8) (0) (7.4) (4.4) (1.5) (4.4)
(6.6) (0) (5.8) (3.6) (2.9) (2.9)
CPB, cardiopulmonary bypass; CVP, central venous pressure; IE, inotropic equivalents; iNO, inhaled nitric oxide; LVFRI, left ventricular flow rate index; MAP, mean arterial pressure; SvO2, mixed venous oxygen saturation. a To qualify for meeting failure criteria, at least 2 criteria (except for death or failure to wean from CPB) must be met. Each patient may be counted in multiple categories.
iNO, inhaled nitric oxide.
failure to wean from CPB may broaden the criteria, the intention was to count any death within 48 hours as RV failure, because if the patient died, it would not be possible to determine if that patient would have had RV failure later. A very conservative assumption was made that any early death was RV-related. No patients actually met the criterion of death. Two deaths occurred during the 48hour period; both patients had already met failure criteria and were already counted. RVD developed in all but 2 patients within approximately 8 hours after LVAD placement (Figure 3). An unexpected finding was the shorter, but non-significant, duration of mechanical ventilation in iNO-treated patients, despite a 23.0% crossover rate. Although this trend
was consistent across centers and sex, it did not translate into a reduced duration of ICU or hospital stay. We also did not find any difference in the 28-day survival rate. The study, however, was designed to protect patients from clinical harm, and differences in survival or other irreversible outcomes were potentially avoided by crossing to openlabel treatment as soon as RVD criteria were met. Therapy with iNO appeared to be safe and well tolerated in this critically ill patient population. No suggestion of increased tendency toward bleeding was evident, despite the use of relatively high iNO doses.
Study limitations This double-blind, placebo-controlled trial allowed rapid cross over to open-label iNO to preserve patient safety. Although ethically necessary, it complicates interpretation of study results. Of the 150 patients, 35 (23.0%) crossed over to open-label iNO, including 20 of 77 (26.0%) in the
Time to crossover to open label in the intent-to-treat population. iNO, inhaled nitric oxide.
876 Table 5
The Journal of Heart and Lung Transplantation, Vol 30, No 8, August 2011 Primary and Secondary Outcome Measures in the Intent-to-Treat Population
Patients meeting RVD criteria ⱕ48 hours No. of total (%) 95% CI Males, No. (%) Females, No. (%) PVRI ⬍270.5 dyne/sec/cm–5 PVRI ⱖ270.5 dyne/sec/cm–5 Days on mechanical ventilationa Mean (SD) Median (range) No. of ICU daysb Mean (SD) Median (range) No. of total hospital daysc Mean (SD) Median (range) Quantity of blood products used Mean, ml (SD) Patients requiring RRT, No. (%)d Non-survival at Day 28, No. (%) Patients needing RVAD by Day 28, No. (%)
7/73 (9.6) 2.8–16.3 7/64 (10.9) 0/9 (0.0) 6/51 (11.8) 1/7 (14.3) 70 5.37 (7.72) 2.0 (1–30) 60 20.52 (32.31) 11.0 (3–194) 58 40.57 (32.19) 32.0 (11–194) 73 4,232 (4675) 10/71 (14.1) 8/71 (11.3) 4/71 (5.6)
12/77 (15.6) 7.5–23.7 7/65 (10.8) 5/12 (41.7) 6/48 (12.5) 5/7 (71.4) 67 11.10 (24.81) 3.0 (0–160) 58 19.90 (24.38) 9.0 (3–115) 58 40.76 (29.41) 31.5 (10–156) 77 4,885 (7760) 8/70 (11.4) 8/70 (11.4) 7/70 (10.0)
⬎0.99 0.045 ⬎0.99 0.103 0.077
0.226 0.637 0.924 0.468
ICU, intensive care unit; iNO, inhaled nitric oxide; PVRI, pulmonary vascular resistance index; RRT, renal replacement therapy; RVAD, right ventricular assist device; RVD, right ventricular dysfunction; SD, standard deviation. a Intent-to-treat population, extubated before study end (Day 28). b Intent-to-treat population, discharged from ICU before study end (Day 28). c Intent-to-treat population, discharged from hospital before study end (Day 28). d Includes those patients in the intent-to-treat population for whom RRT information was available.
placebo group. Hence, any simple comparisons of outcomes or events, other than primary outcome variables, are probably confounded with regard to treatment effect. All secondary outcome variables most likely underestimate the true Table 6
Adverse Events in the Safety Population
Event Any adverse eventsa Blood and lymphatic system disorders Cardiac disorders General disorders Procedural complications Investigationsb Metabolic disorders Others Serious adverse eventsc Thrombocytopenia Right ventricular failure Hemorrhage Procedural complication Hemothorax Fatal adverse eventsc
iNO (n ⫽ 69) No. (%)
Placebo (n ⫽ 68) No. (%)
2 5 2 4 3 2 5 7 0 3 3 1 0 1
(2.9) (7.2) (2.9) (5.8) (4.3) (2.9) (7.2) (10.1) (0) (4.3) (4.3) (1.4) (0) (1.4)
4 6 0 6 2 0 5 11 1 3 6 0 1 1
(5.9) (8.8) (0) (8.8) (2.9) (0) (7.4) (16.2) (1.5) (4.4) (8.8) (0) (1.5) (1.5)
iNO, inhaled nitric oxide. a During blinded study drug treatment. b Represents abnormal diagnostic testing levels or central venous pressure. c During blinded study drug treatment and open-label iNO period.
effect size. This may explain our observation of no betweengroup differences for death or significant morbidities. Furthermore, 15 of 150 patients (6 iNO, 9 placebo) from 1 study center crossed over before meeting the studydefined criteria. This single-site crossover occurrence was a result of unwillingness to wait until study-defined failure criteria were met; the rationale for this decision cannot be assumed to be a result of RVD outside of the specified criteria. However, although this does not change the number of events contributing to the primary end point, it is possible that RVD may have developed in the patients from this crossover group. Although iNO benefit on the physiologic mechanisms of RVD has been shown in small studies, the size of clinical treatment effect was not previously reported, and the rate of RVD has been decreasing over time. The original sample size of 110 patients assumed a placebo event rate of 50%, which was reduced by 50% under iNO treatment. During blinded surveillance, a combined event rate of approximately 15% was observed, and the sample size was arbitrarily increased from 110 to 150 patients. The final event rate was 12.7%. Consequently, the study had⬍56% power to detect a significant difference. To design a trial based on the observed event rates, treatment arms of ⬎500 patients would be required to achieve appropriate powering for the study. In addition, 13 individuals (8.7%) never received study treatment and are included in the ITT analysis. These patients may well have received iNO or other pulmonary vasodilators outside the study protocol. Finally, another 18 participants (12%, 9 in each group) crossed over to open-
Potapov et al.
Use of Inhaled Nitric Oxide After LVAD Placement
Kaplan-Meier plot of time to meeting right ventricular dysfunction criteria in the intent-to-treat population.
label iNO during the 48-hour study treatment period at the discretion of an attending physician, without meeting study criteria, potentially further confounding the results.
Conclusions Use of iNO at 40 ppm given before separation from CPB did not reach statistical significance for the primary end point of reduction in RVD incidence. No statistically significant difference was found for secondary variables, including time on mechanical ventilation, ICU or hospital stay, and the need for RVAD after LVAD placement. Given the inherent limitations of any randomized trial in this setting, and with the complexity and extreme hemodynamic variability typically encountered early after coming off CPB, conducting a rigorous test of strategies during this period may prove difficult.
Disclosure statement This study was sponsored by and statistical analysis of the data was conducted by Ikaria Inc, Clinton, New Jersey. General editorial support (eg, format, style) for this article was provided by Peloton Advantage LLC and funded by INO Therapeutics LLC, a subsidiary of Ikaria Inc. All authors were responsible for data analysis and interpretation, as well as the development and final approval of the paper.
Conception and design of the study were conducted by the Steering Committee members listed in Appendix 1. Drs Meyer, Zucker, Swaminathan, and Marczin have received honoraria as Advisory Board consultants to the sponsor of this trial, and Drs Swaminathan, Ramsay, Marczin, and Zucker have received research grants from the sponsor company. Drs Chittuluru and Baldassarre are employees of the sponsor company. Drs Potapov, El Banayosy, Diehl, Veynovich, Kukucka, Gregoric, Gromann, and Hetzer do not have a financial relationship with a commercial entity that has an interest in the subject of the presented manuscript or other conflicts of interest to disclose.
Appendix 1: Steering Committee Mark J. Zucker, MD, Newark Beth Israel Medical Center, Newark, New Jersey; Dan Meyer, MD, University of Texas Southwestern/St. Paul Medical Center, Dallas, Texas; Michael Ramsay, MD, Baylor University Medical Center, Dallas, Texas; Nandor Marczin, MD, PhD, Imperial College London, UK; and James S. Baldassarre, MD, Ikaria, Clinton, New Jersey.
Appendix 2: Investigators United States: Mark J. Zucker, MD, Newark Beth Israel Medical Center, Newark, New Jersey; Madhav Swaminathan, MD, Duke University Medical Center, Durham, North
The Journal of Heart and Lung Transplantation, Vol 30, No 8, August 2011
Carolina; Dan Meyer, MD, University of Texas Southwestern/St. Paul Medical Center and Michael Ramsay, MD, Baylor University Medical Center, Dallas, Texas; Bryan Veynovich, DO, Allegheny General Hospital, Pittsburgh, Pennsylvania; Igor D. Gregoric, MD, Texas Heart Institute, Houston, Texas; and Aly El Banayosy, MD, Penn State Milton S. Hershey Medical Center, Hershey, Pennsylvania. Germany: Roland Hetzer, MD, PhD, Marian Kukucka, MD, Tom W. Gromann, MD, and Evgenij Potapov, MD, PhD, Deutsches Herzzentrum Berlin, Berlin; and Christoph Diehl, MD, Herz- und Diabeteszentrum Nordrhein-Westfalen, Bad Oeynhausen.
References 16. 1. Rose EA, Gelijns AC, Moskowitz AJ, et al. Long-term mechanical left ventricular assistance for end-stage heart failure. N Engl J Med 2001; 345:1435-43. 2. Deng MC, Edwards LB, Hertz MI, et al. Mechanical circulatory support device database of the International Society for Heart and Lung Transplantation: third annual report—2005. J Heart Lung Transplant 2005;24:1182-7. 3. Santambrogio L, Bianchi T, Fuardo M, et al. Right ventricular failure after left ventricular assist device insertion: preoperative risk factors. Interact Cardiovasc Thorac Surg 2006;5:379-82. 4. Dang NC, Topkara VK, Mercando M, et al. Right heart failure after left ventricular assist device implantation in patients with chronic congestive heart failure. J Heart Lung Transplant 2006;25:1-6. 5. Matthews JC, Koelling TM, Pagani FD, Aaronson KD. The right ventricular failure risk score: a pre-operative tool for assessing the risk of right ventricular failure in left ventricular assist device candidates. J Am Coll Cardiol 2008;51:2163-72. 6. Puwanant S, Hamilton KK, Klodell CT, et al. Tricuspid annular motion as a predictor of severe right ventricular failure after left ventricular assist device implantation. J Heart Lung Transplant 2008;27:1102-7. 7. Fitzpatrick JR 3rd, Frederick JR, Hsu VM, et al. Risk score derived from pre-operative data analysis predicts the need for biventricular mechanical circulatory support. J Heart Lung Transplant 2008;27:1286-92. 8. Kavarana MN, Pessin-Minsley MS, Urtecho J, et al. Right ventricular dysfunction and organ failure in left ventricular assist device recipients: a continuing problem. Ann Thorac Surg 2002;73:745-50. 9. Potapov EV, Stepanenko A, Dandel M, et al. Tricuspid incompetence and geometry of the right ventricle as predictors of right ventricular function after implantation of a left ventricular assist device. J Heart Lung Transplant 2008;27:1275-81. 10. Kormos RL, Teuteberg JJ, Pagani FD, et al. Right ventricular failure in patients with the HeartMate II continuous-flow left ventricular assist
device: incidence, risk factors, and effect on outcomes. J Thorac Cardiovasc Surg 2010;139:1316-24. Lee S, Kamdar F, Madlon-Kay R, et al. Effects of the HeartMate II continuous-flow left ventricular assist device on right ventricular function. J Heart Lung Transplant 2010;29:209-15. Argenziano M, Choudhri AF, Moazami N, et al. Randomized, doubleblind trial of inhaled nitric oxide in LVAD recipients with pulmonary hypertension. Ann Thorac Surg 1998;65:340-5. Hare JM, Shernan SK, Body SC, Graydon E, Colucci WS, Couper GS. Influence of inhaled nitric oxide on systemic flow and ventricular filling pressure in patients receiving mechanical circulatory assistance. Circulation 1997;95:2250-3. Wagner F, Dandel M, Gunther G, et al. Nitric oxide inhalation in the treatment of right ventricular dysfunction following left ventricular assist device implantation. Circulation 1997;96:II-6. Wagner FD, Buz S, Knosalla C, Hetzer R, Hocher B. Modulation of circulating endothelin-1 and big endothelin by nitric oxide inhalation following left ventricular assist device implantation. Circulation 2003; 108 (Suppl 1):II278-84. Ardehali A, Laks H, Levine M, et al. A prospective trial of inhaled nitric oxide in clinical lung transplantation. Transplantation 2001;72: 112-5. Fattouch K, Sbraga F, Bianco G, et al. Inhaled prostacyclin, nitric oxide, and nitroprusside in pulmonary hypertension after mitral valve replacement. J Card Surg 2005;20:171-6. Fattouch K, Sbraga F, Sampognaro R, et al. Treatment of pulmonary hypertension in patients undergoing cardiac surgery with cardiopulmonary bypass: a randomized, prospective, double-blind study. J Cardiovasc Med 2006;7:119-23. Miller OI, Tang SF, Keech A, Pigott NB, Beller E, Celermajer DS. Inhaled nitric oxide and prevention of pulmonary hypertension after congenital heart surgery: a randomised double-blind study. Lancet 2000;356:1464-9. Rajek A, Pernerstorfer T, Kastner J, et al. Inhaled nitric oxide reduces pulmonary vascular resistance more than prostaglandin E(1) during heart transplantation. Anesth Analg 2000;90:523-30. Solina A, Papp D, Ginsberg S, et al. A comparison of inhaled nitric oxide and milrinone for the treatment of pulmonary hypertension in adult cardiac surgery patients. J Cardiothorac Vasc Anesth 2000; 14:12-7. Oz MC, Ardehali A. Collective review: perioperative uses of inhaled nitric oxide in adults. Heart Surg Forum 2004;7:E584-9. Costard-Jackle A, Fowler MB. Influence of preoperative pulmonary artery pressure on mortality after heart transplantation: testing of potential reversibility of pulmonary hypertension with nitroprusside is useful in defining a high risk group. J Am Coll Cardiol 1992;19:48-54. Aaronson KD, Patel H, Pagani FD. Patient selection for left ventricular assist device therapy. Ann Thorac Surg 2003;75:S29-35. Potapov EV, Sodian R, Loebe M, Drews T, Dreysse S, Hetzer R. Revascularization of the occluded right coronary artery during left ventricular assist device implantation. J Heart Lung Transplant 2001; 20:918-22.