Pediatric Pulmonology 50:588–595 (2015)
Iloprost for Children With Pulmonary Hypertension After Surgery to Correct Congenital Heart Disease Zhuoming Xu, PhD,1* Limin Zhu, MD,1 Xinrong Liu, MD,1 Xiaolei Gong, William Gattrell, PhD,2,3 and Jinfen Liu, MD1
1 MD,
Summary. Congenital heart disease (CHD) can cause pulmonary hypertension (PH) in children, and surgery to correct CHD may be complicated by postoperative pulmonary hypertensive crises (PHC). Clinical data regarding the use of inhaled iloprost to treat children with PH are scarce. Our aim was to determine the efficacy and safety of iloprost in children with PH following surgery to correct CHD. This was a randomized, placebo-controlled study of 22 children (median age 7 months) undergoing surgery to achieve biventricular repair. The combined clinical endpoint was a decrease of more than 20% in the ratio of systolic pulmonary arterial pressure to systolic arterial pressure or pulmonary resistance to systemic resistance, with no PHC or death. Patients were randomized to receive low-dose iloprost (30 ng/kg/min), high-dose iloprost (50 ng/kg/min), or placebo, for 10 min every 2 hr in the first 48 hr after surgery. PHC were experienced by two patients who received placebo and one patient treated with high-dose iloprost. The combined clinical endpoint was reached by six patients administered low-dose iloprost (P ¼ 0.005) and four administered high-dose iloprost (P ¼ 0.077), compared with none in the placebo group. Patients treated with iloprost showed a significant reduction from baseline in mean pulmonary vascular resistance index (2.2 Wood units, P < 0.05), whereas patients who received placebo showed no significant change. This study supports the use of iloprost to treat children with PH following surgery to correct CHD. Pediatr Pulmonol. 2015;50:588–595. ß 2014 Wiley Periodicals, Inc. Key words: pulmonary hypertensive crisis; synthetic prostacyclin analogue; hemodynamic parameters; safety. Funding source: Bayer Pharma AG.
INTRODUCTION
The prognosis for children with pulmonary hypertension (PH) is poor, and the mean survival time from diagnosis is less than 1 year if the condition is not treated.1,2 PH associated with congenital heart disease (CHD) accounts for almost 50% of cases of PH in children and may be due to increased pulmonary blood flow, increased pulmonary vascular resistance (PVR) or a combination of both.3 Surgery to correct CHD can prevent the development of
1 Shanghai Children’s Medical Center, Shanghai Jiao Tong University School of Medicine, Shanghai, China. 2
Research Evaluation Unit, Oxford PharmaGenesisTM Ltd, Oxford, UK.
3 Department of Mechanical Engineering and Mathematical Sciences, Oxford Brookes University, Oxford, UK.
Conflict of interest: None. Disclosure: The writing of this article was funded in part by Bayer Pharma AG. Dr Gattrell is an employee of Oxford PharmaGenesisTM Ltd, which has received project funding from Bayer Pharma AG.
ß 2014 Wiley Periodicals, Inc.
pulmonary vascular disease; however, the decision to operate is not necessarily straightforward.4,5 In some patients, PH becomes irreversible following surgery, and the prognosis can be worse than if the procedure had not been carried out.6 Furthermore, pulmonary vascular reactivity is heightened in the postoperative setting, and vasospastic stimuli may result in dramatic increases in pulmonary resistance.7 Such episodes, known as pulmonary hypertensive crises (PHC), are significant contributors to morbidity and mortality in children with PH.8,9
Correspondence to: Zhuoming Xu, Cardiac Intensive Care Unit, Department of Thoracic and Cardiovascular Surgery, Shanghai Children’s Medical Center, Shanghai Jiao Tong University School of Medicine, 1678 Dongfang Road, Shanghai 200127, China. E-mail: [email protected] Received 23 May 2013; Revised 6 January 2014; Accepted 3 February 2014. DOI 10.1002/ppul.23032 Published online 9 March 2014 in Wiley Online Library (wileyonlinelibrary.com).
Iloprost in Children After Cardiac Surgery
The management of PH following cardiac repair is an important clinical consideration, but therapeutic options for children are limited. Clinical evidence continues to emerge of the successful use of inhaled nitric oxide (iNO) to improve the hemodynamic parameters and postoperative outcomes of patients.10 However, administration of iNO can result in the down-regulation of endogenous NO production and increased levels of the vasoconstrictor endothelin-1, contributing to the rebound PH that can occur following discontinuing use of iNO.11 The use of iNO is also associated with the risk of exposure to nitrogen dioxide and adverse events, such as methemoglobinemia. Furthermore, a complex and costly delivery system is required to administer iNO, meaning that it is not available in all parts of the world.12 Iloprost (Ventavis1, Bayer Pharma AG, Berlin, Germany) is a stabilized prostacyclin analogue that may have potential in the acute treatment of postoperative PH in children and neonates, but clinical data are scarce.13 Unlike iNO, iloprost can be administered as an intravenous infusion or can be inhaled by using a nebulizer, which can be incorporated into a ventilatory circuit for use in the intensive care setting. The aim of this study was to determine the efficacy, safety, and hemodynamic effects of inhaled iloprost in the treatment of children with PH following surgery to correct CHD. MATERIALS AND METHODS
The protocol for this single-center, randomized, placebo-controlled, single-blind study was approved by the Institutional Health Research Ethics Board of the Shanghai Children’s Medical Center, Shanghai Jiao Tong University School of Medicine, Shanghai, China. The parents or guardians of the children had given written informed consent to participate in the study. The severity of PH may be classified according to the ratio of systolic pulmonary arterial pressure to systolic arterial pressure (Pp/Ps) or pulmonary resistance to systemic resistance (Rp/Rs). Pulmonary vascular disease is considered moderate if Pp/Ps or Rp/Rs is between 0.45 and 0.75, and severe if Pp/Ps or Rp/Rs is more than 0.75.14 Criteria for Patient Inclusion Prior to Surgery
Children with CHD and associated PH risk factors undergoing surgery to achieve biventricular repair were candidates for inclusion in the study, and were required to fulfill at least two of the criteria below. In patients with left–right cardiac shunt, arterial oxygen saturation was below 93%. There was evidence of right ventricular hypertrophy and right atrial dilatation without right ventricular outflow tract obstruction, as determined by electrocardiography.
589
There was evidence of pulmonary arterial trunk dilatation, right ventricular enlargement and “pruning” of the peripheral blood vessels, as determined by chest X-ray. For patients with ventricular and aortic level bidirectional shunt or left–right cardiac shunt, mean pulmonary arterial pressure (mPAP) was being at least 25 mmHg, as calculated by tricuspid or pulmonary valve regurgitation velocity using cardiac echocardiography. Pp/Ps was above 0.75. The ratio of pulmonary flow to systemic flow (Qp/Qs) was below 1.5. PVR was greater than 9 Wood units/m2 or Rp/Rs was above 0.5. Criteria for Patient Exclusion Prior to Surgery
Exclusion criteria before surgery were body weight below 2 kg, premature birth (36 weeks after conception), renal dysfunction (serum creatinine 1.5 mg/dl 48 hr before surgery), and low cardiac output syndrome or hypotension on arrival at the intensive care unit. Inclusion and Exclusion Criteria After Surgery
Following corrective surgery for CHD, patients were eligible for inclusion in the study if mPAP was at least 25 mmHg, as calculated by tricuspid or pulmonary valve regurgitation velocity, or if Pp/Ps was above 0.5. However, individuals were excluded from the study if they still had cardiac deficiencies that were associated with intracardiac shunts and severe mitral or tricuspid valve regurgitation, if low cardiac output due to severe arrhythmia was present, or if the platelet count was below 50 109/l and obvious bleeding was observed. Procedures
A pulse-induced continuous cardiac output (PICCO) catheter was placed in the femoral artery after stabilization of the patient’s condition. Blood transfusion and administration of drugs selectively reducing pulmonary circulation resistance were stopped 30 min before study treatment was initiated. Arterial blood gas parameters were maintained at the following levels: carbon dioxide tension 30–35 mmHg; arterial oxygen tension above 95 mmHg; and pH 7.5–7.55. Based on clinical need, sedatives, muscle relaxants, non-pulmonary selective vasoactive drugs and hyperventilation could be used to prevent excessive increases in pulmonary arterial pressure. Using a random numbers table, patients were randomized in a 1:1:1 ratio to receive low-dose iloprost (30 ng/kg/ min), high-dose iloprost (50 ng/kg/min) or placebo. This was the dose of iloprost calculated to reach the patient. Pediatric Pulmonology
590
Xu et al.
Study Drug
Study treatment was administered in the first 48 hr after admission to the cardiac intensive care unit following corrective surgery for CHD. Iloprost (2 ml) was given every 120 min, up to 14 times over the study period, via an ultrasonic atomizer device (multisonic1 InfraControl, Schill GmbH, Probstzella, Germany) attached to the patient’s ventilator; each inhalation period lasted 10 min. The same model of ventilator (SERVO-i, Maquet Critical Care AB, Solna, Sweden) was used for all patients. After 48 hr, patients were weaned off the ventilators. Outcome Measures
The following parameters were determined before, immediately after and 30 min after each administration of study treatment: heart rate (HR), systolic arterial pressure (SAP), diastolic arterial pressure (DAP), mean arterial pressure (MAP), systolic pulmonary arterial pressure (sPAP); diastolic pulmonary arterial pressure (dPAP), mPAP, central venous pressure (CVP), left atrial pressure (LAP), cardiac index (CI), Pp/Ps, Rp/Rs, PVR index (PVRI), systemic vascular resistance index (SVRI), and transpulmonary pressure gradient (TPG). The primary combined clinical endpoint was a decrease of more than 20% in Pp/Ps or Rp/Rs with no PHC or death. The secondary clinical endpoints were improvements in hemodynamic parameters. In the event of Pp/Ps being higher than 0.5 for more than 30 min, or Pp/Ps being above 1, administration of study treatment was immediately stopped and other therapies, such as iNO, were employed. Statistical Analysis
All patients were included in the analysis of the primary and secondary clinical endpoints. Baseline was defined as the time point immediately before inhalation of the first study dose, and endpoint was immediately after inhalation of the final study dose. The co-primary endpoints were analyzed using a paired t-test, an unpaired t-test or an analysis of variance, as appropriate. To control for type I errors, the Bonferroni method was used, when appropriate. Changes from baseline in hemodynamic parameters were analyzed with either the Bonferroni or Student–Newman–Keuls test. A P value below 0.05 was considered to be statistically significant. RESULTS Baseline Characteristics
During the period January 2010 to December 2010, 22 patients (median age 7 months) were enrolled in the study (Table 1). Most (17) were male. There was no significant difference between the placebo and iloprost groups in Pediatric Pulmonology
terms of age or weight of the patients. Individuals presented with a range of CHDs, and more than one defect was diagnosed in most patients. Adverse Events, Interventions, and Outcomes
No deaths occurred during the study. Pulmonary hypertensive crises were experienced by two patients who received placebo (28.6%) and one patient who was treated with high-dose iloprost (6.7%) (Table 2). These episodes were successfully managed with iNO. There were two cases of equipment malfunction. In one instance, the PICCO device broke down; in the other, there was extubation of the endotracheal tube. Two patients required surgery to correct unresolved vascular defects. Six experienced thrombocytopenia, and in two severe cases the patients received platelet transfusions. The number of drug administrations per patient did not differ significantly between treatment groups. Primary Clinical Endpoints Change in Pp/Ps or Rp/Rs From Baseline
In the placebo group, mean Pp/Ps increased by 0.15 (32.3%) compared with baseline (P ¼ 0.032), to reach 0.62 by the end of the study (Fig. 1a, c). By contrast, those treated with low-dose iloprost showed a significant reduction from baseline in mean Pp/Ps of 0.10 (22.1%) to a final value of 0.34 (P ¼ 0.032) (Fig. 1a, c). Compared with baseline, patients who received high-dose iloprost showed a reduction in mean Pp/Ps of 0.04 (9.5%), although this was not statistically significant (Fig. 1a). In the combined group of individuals who received iloprost, mean Pp/Ps decreased to 0.34, a reduction of 0.06 (16.0%) from baseline (P ¼ 0.016) (Fig. 1a, c). In patients treated with placebo, there was an increase in mean Rp/Rs of 0.18 (49.4%) compared with baseline (P ¼ 0.041), to reach 0.53 (Fig. 1b, d). By contrast, mean Rp/Rs decreased from baseline by 0.14 (38.5%) to 0.22 in patients treated with low-dose iloprost (P ¼ 0.034). Patients treated with high-dose iloprost showed a 0.04 (15.5%) decrease from baseline in mean Rp/Rs, to 0.22 (Fig. 1b, d); however, this reduction was not statistically significant. In the combined group of patients who were treated with iloprost, mean Rp/Rs significantly decreased from baseline by 0.09 (28.1%), to 0.22 (P ¼ 0.026). Change in Pp/Ps or Rp/Rs Compared With Placebo
Immediately after inhalation of the final dose of study treatment, mean Pp/Ps was significantly lower in patients who received low-dose or high-dose iloprost than in those who received placebo (P ¼ 0.046 and P ¼ 0.039, respectively) (Fig. 1c). Mean Pp/Ps decreased by 0.27 in patients treated with low-dose iloprost and by 0.28 in those treated with high-dose iloprost compared with placebo, a difference of 44.5% and 46.1%, respectively.
Iloprost in Children After Cardiac Surgery
591
TABLE 1— Demographics of Patients and Diagnoses of Congenital Heart Disease
Patient number
Sex
Age (months)
Weight (kg)
Placebo group 4 5 7
M F F
0.4 63 6
3.5 12 5.5
12
M
2
3.9
13
M
3
4.6
21 22
M F
13 2
8 4
Mean (SD) 12.8 (20.9) Low-dose iloprost (30 ng/kg/min) 1 M 19 9 F 147 10 M 5 11 F 1
12 23 4.5 2.5
M
47
12
16 18
M M
6 2
7.5 4.8
32.4 (49.1) (50 ng/kg/min) M 73 M 12 M 11
of the great arteries; atrial septal defect Ventricular septal defect; patent ductus arteriosus D-transposition of the great arteries; ventricular septal defect Interrupted aortic arch (type B); aortopulmonary window; patent ductus arteriosus D-transposition of the great arteries; ventricular septal defect; atrial septal defect Total anomalous pulmonary venous connection Interrupted aortic arch (type A); ventricular septal defect; patent foramen ovale; patent ductus arteriosus
18 8 8.5
M
9
6
14 17
M M
7 1.6
4.5 5.6
M M
72 7 24.1 (28.1) 0.596
15.5 6 9.0 (4.7) 0.403
14 12 4 9 12 12 13
10.9 (3.1) Aortopulmonary window Complete atrioventricular canal defect Taussig–Bing anomaly D-transposition of the great arteries; ventricular and atrial septal defect; patent ductus arteriosus; coarctation of the aorta Interrupted aortic arch (type B); ventricular septal defect; patent foramen ovale Persistent truncus arteriosus; ventricular septal defect D-transposition of the great arteries; multi-ventricular septal defects; atrial septal defect
9.5 (6.5)
8
19 20 Mean (SD) P-value
D-transposition
5.9 (2.8)
15
Mean (SD) High-dose iloprost 2 3 6
No. of treatment administrations
Diagnoses
12 6 12 11
11 13 11 10.9 (2.1)
Aortopulmonary window Taussig-Bing anomaly D-transposition of the great arteries; ventricular septal defect D-transposition of the great arteries; ventricular septal defect Persistent truncus arteriosus; ventricular septal defect Interrupted aortic arch (type A); ventricular septal defect; patent foramen ovale; patent ductus arteriosus Ventricular septal defect; atrial septal defect Total anomalous pulmonary venous connection
11 7 13 12 14 12
6 11 10.7 (2.6) 0.994
SD, standard deviation.
In patients treated with either low-dose or high-dose iloprost, mean Rp/Rs decreased by 0.31 compared with placebo over the study period; however, this did not reach statistical significance (P ¼ 0.08) (Fig. 1d).
the individuals treated with placebo. The odds ratio could not be calculated because no patient in the placebo group reached the combined clinical endpoint. Secondary Clinical Endpoints
Combined Clinical Endpoint
The combined clinical endpoint of a decrease of more than 20% in Pp/Ps or Rp/Rs with no PHC or death was reached by six of the patients (85.7%) administered lowdose iloprost (P ¼ 0.005) and four (50.0%) administered high-dose iloprost (P ¼ 0.077), compared with none of
In patients who received placebo, there was no significant change in the measured hemodynamic parameters after inhalation of the final dose of study treatment compared with values recorded before inhalation of the first study treatment (Table 3). Compared with baseline, there was a significant reduction in mean HR, Pediatric Pulmonology
592
Xu et al.
TABLE 2— Adverse Events, Complications, Interventions, and Outcomes
Patient number Placebo group 4 5 7 12
Adverse event or complication
Minimal decrease in BPC Minimal decrease in BPC Pulmonary hypertensive crisis Pulmonary hypertensive crisis Right bronchus compressed owing to RPA overlaying between AAo and DAo Low-dose iloprost (30 ng/kg/min) 9 PICCO device broke down Severe decrease in BPC 11 Residual coarctation of aorta 16 Severe decrease in BPC High-dose iloprost (50 ng/kg/min) 3 Pulmonary hypertensive crisis Minimal decrease in BPC 14 Moderate decrease in BPC 19 Extubation of endotracheal tube
Time after surgery
Intervention
Outcome
1 day 3 days 18 hr (fifth inhalation) 36 hr (ninth inhalation) 7 days
None None iNO iNO Surgery
Alleviated after 3 days Alleviated after 3 days Alleviated Alleviated Resolved
19 hr 3 days 2 days 3 days
Conventional therapy Transfusion of 1 unit of platelets Surgery Transfusion of 1 unit of platelets
Resolved Alleviated after 8 days Resolved Alleviated after 2 days
30 hr (eighth inhalation) 2 days 3 days 22 hr
iNO None None Observation
Alleviated Alleviated after 1 day Alleviated after 5 days Successful cessation of ventilator use
AAo, ascending aorta; BPC, blood platelet count; DAo, descending aorta; iNO, inhaled nitric oxide; PICCO, pulse-induced continuous cardiac output; RPA, right pulmonary artery.
sPAP, dPAP, and mPAP in patients who received low-dose iloprost. Mean PVRI was also reduced compared with baseline, although this did not reach statistical significance (P ¼ 0.052). There were no significant changes from baseline to endpoint in the hemodynamic parameters of patients who received high-dose iloprost. However, the combined group of patients treated with iloprost showed statistically significant reductions in mean values of sPAP, dPAP, mPAP, and PVRI. DISCUSSION
There are very limited data available on the treatment of children with iloprost, particularly in the postoperative setting, and most published studies have examined only small numbers of patients.13 In many pediatric intensive care units, iNO is a key pharmacotherapy used in the management of PH, and several studies have compared the efficacy of iloprost with this treatment.10 In a recent study by Kirbas et al., children who underwent surgery for CHD showed a significant reduction in PAP and Pp/Ps from preoperative values, and there were no significant differences between iNO and iloprost in terms of the efficacy and rates of adverse events.15 In a study by Loukanov et al., after cardiopulmonary bypass surgery there were no significant differences in the frequency of PHC, mPAP values or duration of mechanical ventilation in children treated with either iNO or aerosolized iloprost.16 Although iloprost and iNO act via different pathways (production of cyclic adenosine monophosPediatric Pulmonology
phate and cyclic guanosine monophosphate, respectively), the combination of both agents gave no additional efficacy compared with each agent separately.17 However, administration of iloprost via nebulization is straightforward, whereas a complex delivery system is required for iNO. Furthermore, daily administration of iNO is reported to be as much as twenty times more expensive than inhaled iloprost.18 Only one published placebo-controlled study has examined the use of iloprost in children following surgery to correct CHDs.19 The absolute values for the hemodynamic parameters measured in the study were not presented, and caution is advised in making direct comparisons with the present study; however, in keeping with our study, treatment with iloprost significantly reduced mPAP, Pp/Ps and Rp/Rs compared with baseline, whereas significant increases in these values were seen in patients treated with placebo. Although a rise in these hemodynamic measures was observed acutely in patients treated with placebo, normal pulmonary vascular pressure is usually restored within 2–3 months of surgery and patients do not experience chronic PH.19 In our study, PHC occurred in two patients who received placebo, and in one patient who received the equivalent of 4 mg of iloprost per inhalation. The latter event occurred during the eighth inhalation of iloprost after surgery; therefore, the possibility of a causative role for the high-dose study drug cannot be excluded. There are a few literature reports of pulmonary vasodilators causing an increase in PVR. In an evaluation of the acute
Iloprost in Children After Cardiac Surgery
a
b
0.4
593
0.4
*
0.2 0.1 *
*
0 –0.1 –0.2
Mean change in Rp/Rs from baseline
Mean change in Pp/Ps from baseline
* 0.3
–0.3 Low-dose iloprost
High-dose iloprost
0.1 * *
0 –0.1 –0.2
Combined iloprost
Placebo
d
1.0
0.9
0.8
0.8
0.7 0.6 0.5 ††
0.4
††
††
0.3 0.2
Low-dose iloprost
High-dose iloprost
Combined iloprost
1.0
0.9 Mean Rp/Rs at end of study
Mean Pp/Ps at end of study
0.2
–0.3 Placebo
c
0.3
0.7 0.6 0.5 0.4
†
†
†
0.3 0.2 0.1
0.1
0
0 Placebo
Low-dose iloprost
High-dose iloprost
Combined iloprost
Placebo
Low-dose iloprost
High-dose iloprost
Combined iloprost
Fig. 1. Mean change in (a) Pp/Ps and (b) Rp/Rs from baseline to study end, and (c) Pp/Ps and (d) Rp/Rs values immediately after inhalation of the final dose of study treatment. P < 0.05 versus baseline; †P ¼ 0.08 versus placebo; ††P < 0.05 versus placebo. Statistical analysis was conducted using the paired sample t-test. High-dose iloporost, 50 ng/kg/min; low-dose iloprost, 30 ng/kg/ min; Pp/Ps, ratio of systolic pulmonary arterial pressure to systolic arterial pressure; Rp/Rs, ratio of pulmonary resistance to systemic resistance.
hemodynamic effects of iloprost and iNO in adult patients with PH, iNO treatment increased PVR in 17% of patients, whereas PAP and PVR decreased or remained unchanged in all patients treated with iloprost.20 An isolated case report described a paradoxical increase in PVR associated with iloprost inhalation in an 8-monthold infant.21 The mechanism by which iloprost might cause an increase in PVR remains unclear. In our study, the pulmonary hypertensive crisis occurred in a very young patient and may have resulted from poor tolerability of the high-dose of iloprost in that patient. In a study of 22 children with pulmonary arterial hypertension by Ivy et al., two young children who received 2.5 mg of iloprost per treatment discontinued use due to bronchoconstriction.22 Although the use of iloprost is generally well tolerated in children, controlled clinical studies could prove valuable in establishing the appropriate dose of iloprost in this setting, especially for young children.
In our study, the combined primary endpoint, a decrease in Pp/Ps or Rp/Rs of more than 20% with no PHC or death, was reached by significantly more patients who received low-dose iloprost than those who received placebo. In the combined iloprost group, the mean Pp/Ps decreased to 0.34, close to the normal value (
Iloprost for Children With Pulmonary Hypertension After Surgery to Correct Congenital Heart Disease Zhuoming Xu, PhD,1* Limin Zhu, MD,1 Xinrong Liu, MD,1 Xiaolei Gong, William Gattrell, PhD,2,3 and Jinfen Liu, MD1
1 MD,
Summary. Congenital heart disease (CHD) can cause pulmonary hypertension (PH) in children, and surgery to correct CHD may be complicated by postoperative pulmonary hypertensive crises (PHC). Clinical data regarding the use of inhaled iloprost to treat children with PH are scarce. Our aim was to determine the efficacy and safety of iloprost in children with PH following surgery to correct CHD. This was a randomized, placebo-controlled study of 22 children (median age 7 months) undergoing surgery to achieve biventricular repair. The combined clinical endpoint was a decrease of more than 20% in the ratio of systolic pulmonary arterial pressure to systolic arterial pressure or pulmonary resistance to systemic resistance, with no PHC or death. Patients were randomized to receive low-dose iloprost (30 ng/kg/min), high-dose iloprost (50 ng/kg/min), or placebo, for 10 min every 2 hr in the first 48 hr after surgery. PHC were experienced by two patients who received placebo and one patient treated with high-dose iloprost. The combined clinical endpoint was reached by six patients administered low-dose iloprost (P ¼ 0.005) and four administered high-dose iloprost (P ¼ 0.077), compared with none in the placebo group. Patients treated with iloprost showed a significant reduction from baseline in mean pulmonary vascular resistance index (2.2 Wood units, P < 0.05), whereas patients who received placebo showed no significant change. This study supports the use of iloprost to treat children with PH following surgery to correct CHD. Pediatr Pulmonol. 2015;50:588–595. ß 2014 Wiley Periodicals, Inc. Key words: pulmonary hypertensive crisis; synthetic prostacyclin analogue; hemodynamic parameters; safety. Funding source: Bayer Pharma AG.
INTRODUCTION
The prognosis for children with pulmonary hypertension (PH) is poor, and the mean survival time from diagnosis is less than 1 year if the condition is not treated.1,2 PH associated with congenital heart disease (CHD) accounts for almost 50% of cases of PH in children and may be due to increased pulmonary blood flow, increased pulmonary vascular resistance (PVR) or a combination of both.3 Surgery to correct CHD can prevent the development of
1 Shanghai Children’s Medical Center, Shanghai Jiao Tong University School of Medicine, Shanghai, China. 2
Research Evaluation Unit, Oxford PharmaGenesisTM Ltd, Oxford, UK.
3 Department of Mechanical Engineering and Mathematical Sciences, Oxford Brookes University, Oxford, UK.
Conflict of interest: None. Disclosure: The writing of this article was funded in part by Bayer Pharma AG. Dr Gattrell is an employee of Oxford PharmaGenesisTM Ltd, which has received project funding from Bayer Pharma AG.
ß 2014 Wiley Periodicals, Inc.
pulmonary vascular disease; however, the decision to operate is not necessarily straightforward.4,5 In some patients, PH becomes irreversible following surgery, and the prognosis can be worse than if the procedure had not been carried out.6 Furthermore, pulmonary vascular reactivity is heightened in the postoperative setting, and vasospastic stimuli may result in dramatic increases in pulmonary resistance.7 Such episodes, known as pulmonary hypertensive crises (PHC), are significant contributors to morbidity and mortality in children with PH.8,9
Correspondence to: Zhuoming Xu, Cardiac Intensive Care Unit, Department of Thoracic and Cardiovascular Surgery, Shanghai Children’s Medical Center, Shanghai Jiao Tong University School of Medicine, 1678 Dongfang Road, Shanghai 200127, China. E-mail: [email protected] Received 23 May 2013; Revised 6 January 2014; Accepted 3 February 2014. DOI 10.1002/ppul.23032 Published online 9 March 2014 in Wiley Online Library (wileyonlinelibrary.com).
Iloprost in Children After Cardiac Surgery
The management of PH following cardiac repair is an important clinical consideration, but therapeutic options for children are limited. Clinical evidence continues to emerge of the successful use of inhaled nitric oxide (iNO) to improve the hemodynamic parameters and postoperative outcomes of patients.10 However, administration of iNO can result in the down-regulation of endogenous NO production and increased levels of the vasoconstrictor endothelin-1, contributing to the rebound PH that can occur following discontinuing use of iNO.11 The use of iNO is also associated with the risk of exposure to nitrogen dioxide and adverse events, such as methemoglobinemia. Furthermore, a complex and costly delivery system is required to administer iNO, meaning that it is not available in all parts of the world.12 Iloprost (Ventavis1, Bayer Pharma AG, Berlin, Germany) is a stabilized prostacyclin analogue that may have potential in the acute treatment of postoperative PH in children and neonates, but clinical data are scarce.13 Unlike iNO, iloprost can be administered as an intravenous infusion or can be inhaled by using a nebulizer, which can be incorporated into a ventilatory circuit for use in the intensive care setting. The aim of this study was to determine the efficacy, safety, and hemodynamic effects of inhaled iloprost in the treatment of children with PH following surgery to correct CHD. MATERIALS AND METHODS
The protocol for this single-center, randomized, placebo-controlled, single-blind study was approved by the Institutional Health Research Ethics Board of the Shanghai Children’s Medical Center, Shanghai Jiao Tong University School of Medicine, Shanghai, China. The parents or guardians of the children had given written informed consent to participate in the study. The severity of PH may be classified according to the ratio of systolic pulmonary arterial pressure to systolic arterial pressure (Pp/Ps) or pulmonary resistance to systemic resistance (Rp/Rs). Pulmonary vascular disease is considered moderate if Pp/Ps or Rp/Rs is between 0.45 and 0.75, and severe if Pp/Ps or Rp/Rs is more than 0.75.14 Criteria for Patient Inclusion Prior to Surgery
Children with CHD and associated PH risk factors undergoing surgery to achieve biventricular repair were candidates for inclusion in the study, and were required to fulfill at least two of the criteria below. In patients with left–right cardiac shunt, arterial oxygen saturation was below 93%. There was evidence of right ventricular hypertrophy and right atrial dilatation without right ventricular outflow tract obstruction, as determined by electrocardiography.
589
There was evidence of pulmonary arterial trunk dilatation, right ventricular enlargement and “pruning” of the peripheral blood vessels, as determined by chest X-ray. For patients with ventricular and aortic level bidirectional shunt or left–right cardiac shunt, mean pulmonary arterial pressure (mPAP) was being at least 25 mmHg, as calculated by tricuspid or pulmonary valve regurgitation velocity using cardiac echocardiography. Pp/Ps was above 0.75. The ratio of pulmonary flow to systemic flow (Qp/Qs) was below 1.5. PVR was greater than 9 Wood units/m2 or Rp/Rs was above 0.5. Criteria for Patient Exclusion Prior to Surgery
Exclusion criteria before surgery were body weight below 2 kg, premature birth (36 weeks after conception), renal dysfunction (serum creatinine 1.5 mg/dl 48 hr before surgery), and low cardiac output syndrome or hypotension on arrival at the intensive care unit. Inclusion and Exclusion Criteria After Surgery
Following corrective surgery for CHD, patients were eligible for inclusion in the study if mPAP was at least 25 mmHg, as calculated by tricuspid or pulmonary valve regurgitation velocity, or if Pp/Ps was above 0.5. However, individuals were excluded from the study if they still had cardiac deficiencies that were associated with intracardiac shunts and severe mitral or tricuspid valve regurgitation, if low cardiac output due to severe arrhythmia was present, or if the platelet count was below 50 109/l and obvious bleeding was observed. Procedures
A pulse-induced continuous cardiac output (PICCO) catheter was placed in the femoral artery after stabilization of the patient’s condition. Blood transfusion and administration of drugs selectively reducing pulmonary circulation resistance were stopped 30 min before study treatment was initiated. Arterial blood gas parameters were maintained at the following levels: carbon dioxide tension 30–35 mmHg; arterial oxygen tension above 95 mmHg; and pH 7.5–7.55. Based on clinical need, sedatives, muscle relaxants, non-pulmonary selective vasoactive drugs and hyperventilation could be used to prevent excessive increases in pulmonary arterial pressure. Using a random numbers table, patients were randomized in a 1:1:1 ratio to receive low-dose iloprost (30 ng/kg/ min), high-dose iloprost (50 ng/kg/min) or placebo. This was the dose of iloprost calculated to reach the patient. Pediatric Pulmonology
590
Xu et al.
Study Drug
Study treatment was administered in the first 48 hr after admission to the cardiac intensive care unit following corrective surgery for CHD. Iloprost (2 ml) was given every 120 min, up to 14 times over the study period, via an ultrasonic atomizer device (multisonic1 InfraControl, Schill GmbH, Probstzella, Germany) attached to the patient’s ventilator; each inhalation period lasted 10 min. The same model of ventilator (SERVO-i, Maquet Critical Care AB, Solna, Sweden) was used for all patients. After 48 hr, patients were weaned off the ventilators. Outcome Measures
The following parameters were determined before, immediately after and 30 min after each administration of study treatment: heart rate (HR), systolic arterial pressure (SAP), diastolic arterial pressure (DAP), mean arterial pressure (MAP), systolic pulmonary arterial pressure (sPAP); diastolic pulmonary arterial pressure (dPAP), mPAP, central venous pressure (CVP), left atrial pressure (LAP), cardiac index (CI), Pp/Ps, Rp/Rs, PVR index (PVRI), systemic vascular resistance index (SVRI), and transpulmonary pressure gradient (TPG). The primary combined clinical endpoint was a decrease of more than 20% in Pp/Ps or Rp/Rs with no PHC or death. The secondary clinical endpoints were improvements in hemodynamic parameters. In the event of Pp/Ps being higher than 0.5 for more than 30 min, or Pp/Ps being above 1, administration of study treatment was immediately stopped and other therapies, such as iNO, were employed. Statistical Analysis
All patients were included in the analysis of the primary and secondary clinical endpoints. Baseline was defined as the time point immediately before inhalation of the first study dose, and endpoint was immediately after inhalation of the final study dose. The co-primary endpoints were analyzed using a paired t-test, an unpaired t-test or an analysis of variance, as appropriate. To control for type I errors, the Bonferroni method was used, when appropriate. Changes from baseline in hemodynamic parameters were analyzed with either the Bonferroni or Student–Newman–Keuls test. A P value below 0.05 was considered to be statistically significant. RESULTS Baseline Characteristics
During the period January 2010 to December 2010, 22 patients (median age 7 months) were enrolled in the study (Table 1). Most (17) were male. There was no significant difference between the placebo and iloprost groups in Pediatric Pulmonology
terms of age or weight of the patients. Individuals presented with a range of CHDs, and more than one defect was diagnosed in most patients. Adverse Events, Interventions, and Outcomes
No deaths occurred during the study. Pulmonary hypertensive crises were experienced by two patients who received placebo (28.6%) and one patient who was treated with high-dose iloprost (6.7%) (Table 2). These episodes were successfully managed with iNO. There were two cases of equipment malfunction. In one instance, the PICCO device broke down; in the other, there was extubation of the endotracheal tube. Two patients required surgery to correct unresolved vascular defects. Six experienced thrombocytopenia, and in two severe cases the patients received platelet transfusions. The number of drug administrations per patient did not differ significantly between treatment groups. Primary Clinical Endpoints Change in Pp/Ps or Rp/Rs From Baseline
In the placebo group, mean Pp/Ps increased by 0.15 (32.3%) compared with baseline (P ¼ 0.032), to reach 0.62 by the end of the study (Fig. 1a, c). By contrast, those treated with low-dose iloprost showed a significant reduction from baseline in mean Pp/Ps of 0.10 (22.1%) to a final value of 0.34 (P ¼ 0.032) (Fig. 1a, c). Compared with baseline, patients who received high-dose iloprost showed a reduction in mean Pp/Ps of 0.04 (9.5%), although this was not statistically significant (Fig. 1a). In the combined group of individuals who received iloprost, mean Pp/Ps decreased to 0.34, a reduction of 0.06 (16.0%) from baseline (P ¼ 0.016) (Fig. 1a, c). In patients treated with placebo, there was an increase in mean Rp/Rs of 0.18 (49.4%) compared with baseline (P ¼ 0.041), to reach 0.53 (Fig. 1b, d). By contrast, mean Rp/Rs decreased from baseline by 0.14 (38.5%) to 0.22 in patients treated with low-dose iloprost (P ¼ 0.034). Patients treated with high-dose iloprost showed a 0.04 (15.5%) decrease from baseline in mean Rp/Rs, to 0.22 (Fig. 1b, d); however, this reduction was not statistically significant. In the combined group of patients who were treated with iloprost, mean Rp/Rs significantly decreased from baseline by 0.09 (28.1%), to 0.22 (P ¼ 0.026). Change in Pp/Ps or Rp/Rs Compared With Placebo
Immediately after inhalation of the final dose of study treatment, mean Pp/Ps was significantly lower in patients who received low-dose or high-dose iloprost than in those who received placebo (P ¼ 0.046 and P ¼ 0.039, respectively) (Fig. 1c). Mean Pp/Ps decreased by 0.27 in patients treated with low-dose iloprost and by 0.28 in those treated with high-dose iloprost compared with placebo, a difference of 44.5% and 46.1%, respectively.
Iloprost in Children After Cardiac Surgery
591
TABLE 1— Demographics of Patients and Diagnoses of Congenital Heart Disease
Patient number
Sex
Age (months)
Weight (kg)
Placebo group 4 5 7
M F F
0.4 63 6
3.5 12 5.5
12
M
2
3.9
13
M
3
4.6
21 22
M F
13 2
8 4
Mean (SD) 12.8 (20.9) Low-dose iloprost (30 ng/kg/min) 1 M 19 9 F 147 10 M 5 11 F 1
12 23 4.5 2.5
M
47
12
16 18
M M
6 2
7.5 4.8
32.4 (49.1) (50 ng/kg/min) M 73 M 12 M 11
of the great arteries; atrial septal defect Ventricular septal defect; patent ductus arteriosus D-transposition of the great arteries; ventricular septal defect Interrupted aortic arch (type B); aortopulmonary window; patent ductus arteriosus D-transposition of the great arteries; ventricular septal defect; atrial septal defect Total anomalous pulmonary venous connection Interrupted aortic arch (type A); ventricular septal defect; patent foramen ovale; patent ductus arteriosus
18 8 8.5
M
9
6
14 17
M M
7 1.6
4.5 5.6
M M
72 7 24.1 (28.1) 0.596
15.5 6 9.0 (4.7) 0.403
14 12 4 9 12 12 13
10.9 (3.1) Aortopulmonary window Complete atrioventricular canal defect Taussig–Bing anomaly D-transposition of the great arteries; ventricular and atrial septal defect; patent ductus arteriosus; coarctation of the aorta Interrupted aortic arch (type B); ventricular septal defect; patent foramen ovale Persistent truncus arteriosus; ventricular septal defect D-transposition of the great arteries; multi-ventricular septal defects; atrial septal defect
9.5 (6.5)
8
19 20 Mean (SD) P-value
D-transposition
5.9 (2.8)
15
Mean (SD) High-dose iloprost 2 3 6
No. of treatment administrations
Diagnoses
12 6 12 11
11 13 11 10.9 (2.1)
Aortopulmonary window Taussig-Bing anomaly D-transposition of the great arteries; ventricular septal defect D-transposition of the great arteries; ventricular septal defect Persistent truncus arteriosus; ventricular septal defect Interrupted aortic arch (type A); ventricular septal defect; patent foramen ovale; patent ductus arteriosus Ventricular septal defect; atrial septal defect Total anomalous pulmonary venous connection
11 7 13 12 14 12
6 11 10.7 (2.6) 0.994
SD, standard deviation.
In patients treated with either low-dose or high-dose iloprost, mean Rp/Rs decreased by 0.31 compared with placebo over the study period; however, this did not reach statistical significance (P ¼ 0.08) (Fig. 1d).
the individuals treated with placebo. The odds ratio could not be calculated because no patient in the placebo group reached the combined clinical endpoint. Secondary Clinical Endpoints
Combined Clinical Endpoint
The combined clinical endpoint of a decrease of more than 20% in Pp/Ps or Rp/Rs with no PHC or death was reached by six of the patients (85.7%) administered lowdose iloprost (P ¼ 0.005) and four (50.0%) administered high-dose iloprost (P ¼ 0.077), compared with none of
In patients who received placebo, there was no significant change in the measured hemodynamic parameters after inhalation of the final dose of study treatment compared with values recorded before inhalation of the first study treatment (Table 3). Compared with baseline, there was a significant reduction in mean HR, Pediatric Pulmonology
592
Xu et al.
TABLE 2— Adverse Events, Complications, Interventions, and Outcomes
Patient number Placebo group 4 5 7 12
Adverse event or complication
Minimal decrease in BPC Minimal decrease in BPC Pulmonary hypertensive crisis Pulmonary hypertensive crisis Right bronchus compressed owing to RPA overlaying between AAo and DAo Low-dose iloprost (30 ng/kg/min) 9 PICCO device broke down Severe decrease in BPC 11 Residual coarctation of aorta 16 Severe decrease in BPC High-dose iloprost (50 ng/kg/min) 3 Pulmonary hypertensive crisis Minimal decrease in BPC 14 Moderate decrease in BPC 19 Extubation of endotracheal tube
Time after surgery
Intervention
Outcome
1 day 3 days 18 hr (fifth inhalation) 36 hr (ninth inhalation) 7 days
None None iNO iNO Surgery
Alleviated after 3 days Alleviated after 3 days Alleviated Alleviated Resolved
19 hr 3 days 2 days 3 days
Conventional therapy Transfusion of 1 unit of platelets Surgery Transfusion of 1 unit of platelets
Resolved Alleviated after 8 days Resolved Alleviated after 2 days
30 hr (eighth inhalation) 2 days 3 days 22 hr
iNO None None Observation
Alleviated Alleviated after 1 day Alleviated after 5 days Successful cessation of ventilator use
AAo, ascending aorta; BPC, blood platelet count; DAo, descending aorta; iNO, inhaled nitric oxide; PICCO, pulse-induced continuous cardiac output; RPA, right pulmonary artery.
sPAP, dPAP, and mPAP in patients who received low-dose iloprost. Mean PVRI was also reduced compared with baseline, although this did not reach statistical significance (P ¼ 0.052). There were no significant changes from baseline to endpoint in the hemodynamic parameters of patients who received high-dose iloprost. However, the combined group of patients treated with iloprost showed statistically significant reductions in mean values of sPAP, dPAP, mPAP, and PVRI. DISCUSSION
There are very limited data available on the treatment of children with iloprost, particularly in the postoperative setting, and most published studies have examined only small numbers of patients.13 In many pediatric intensive care units, iNO is a key pharmacotherapy used in the management of PH, and several studies have compared the efficacy of iloprost with this treatment.10 In a recent study by Kirbas et al., children who underwent surgery for CHD showed a significant reduction in PAP and Pp/Ps from preoperative values, and there were no significant differences between iNO and iloprost in terms of the efficacy and rates of adverse events.15 In a study by Loukanov et al., after cardiopulmonary bypass surgery there were no significant differences in the frequency of PHC, mPAP values or duration of mechanical ventilation in children treated with either iNO or aerosolized iloprost.16 Although iloprost and iNO act via different pathways (production of cyclic adenosine monophosPediatric Pulmonology
phate and cyclic guanosine monophosphate, respectively), the combination of both agents gave no additional efficacy compared with each agent separately.17 However, administration of iloprost via nebulization is straightforward, whereas a complex delivery system is required for iNO. Furthermore, daily administration of iNO is reported to be as much as twenty times more expensive than inhaled iloprost.18 Only one published placebo-controlled study has examined the use of iloprost in children following surgery to correct CHDs.19 The absolute values for the hemodynamic parameters measured in the study were not presented, and caution is advised in making direct comparisons with the present study; however, in keeping with our study, treatment with iloprost significantly reduced mPAP, Pp/Ps and Rp/Rs compared with baseline, whereas significant increases in these values were seen in patients treated with placebo. Although a rise in these hemodynamic measures was observed acutely in patients treated with placebo, normal pulmonary vascular pressure is usually restored within 2–3 months of surgery and patients do not experience chronic PH.19 In our study, PHC occurred in two patients who received placebo, and in one patient who received the equivalent of 4 mg of iloprost per inhalation. The latter event occurred during the eighth inhalation of iloprost after surgery; therefore, the possibility of a causative role for the high-dose study drug cannot be excluded. There are a few literature reports of pulmonary vasodilators causing an increase in PVR. In an evaluation of the acute
Iloprost in Children After Cardiac Surgery
a
b
0.4
593
0.4
*
0.2 0.1 *
*
0 –0.1 –0.2
Mean change in Rp/Rs from baseline
Mean change in Pp/Ps from baseline
* 0.3
–0.3 Low-dose iloprost
High-dose iloprost
0.1 * *
0 –0.1 –0.2
Combined iloprost
Placebo
d
1.0
0.9
0.8
0.8
0.7 0.6 0.5 ††
0.4
††
††
0.3 0.2
Low-dose iloprost
High-dose iloprost
Combined iloprost
1.0
0.9 Mean Rp/Rs at end of study
Mean Pp/Ps at end of study
0.2
–0.3 Placebo
c
0.3
0.7 0.6 0.5 0.4
†
†
†
0.3 0.2 0.1
0.1
0
0 Placebo
Low-dose iloprost
High-dose iloprost
Combined iloprost
Placebo
Low-dose iloprost
High-dose iloprost
Combined iloprost
Fig. 1. Mean change in (a) Pp/Ps and (b) Rp/Rs from baseline to study end, and (c) Pp/Ps and (d) Rp/Rs values immediately after inhalation of the final dose of study treatment. P < 0.05 versus baseline; †P ¼ 0.08 versus placebo; ††P < 0.05 versus placebo. Statistical analysis was conducted using the paired sample t-test. High-dose iloporost, 50 ng/kg/min; low-dose iloprost, 30 ng/kg/ min; Pp/Ps, ratio of systolic pulmonary arterial pressure to systolic arterial pressure; Rp/Rs, ratio of pulmonary resistance to systemic resistance.
hemodynamic effects of iloprost and iNO in adult patients with PH, iNO treatment increased PVR in 17% of patients, whereas PAP and PVR decreased or remained unchanged in all patients treated with iloprost.20 An isolated case report described a paradoxical increase in PVR associated with iloprost inhalation in an 8-monthold infant.21 The mechanism by which iloprost might cause an increase in PVR remains unclear. In our study, the pulmonary hypertensive crisis occurred in a very young patient and may have resulted from poor tolerability of the high-dose of iloprost in that patient. In a study of 22 children with pulmonary arterial hypertension by Ivy et al., two young children who received 2.5 mg of iloprost per treatment discontinued use due to bronchoconstriction.22 Although the use of iloprost is generally well tolerated in children, controlled clinical studies could prove valuable in establishing the appropriate dose of iloprost in this setting, especially for young children.
In our study, the combined primary endpoint, a decrease in Pp/Ps or Rp/Rs of more than 20% with no PHC or death, was reached by significantly more patients who received low-dose iloprost than those who received placebo. In the combined iloprost group, the mean Pp/Ps decreased to 0.34, close to the normal value (
