Introduction
Pulmonary hypertension is a major cause of mortality and morbidity, affecting approximately 2 per 1000 live births annually. Reference Steinhorn1 Mortality approaches 10% for infants with moderate-to-severe disease and increases further in premature infants or those with pre-existing cardiac and pulmonary anomalies. Reference Roberts, Fineman and Morin2 Of infants who survive, at least 25% face significant long-term morbidities such as chronic lung disease and neurodevelopmental impairment. Reference Steurer, Baer and Oltman3 Despite advances in neonatal ventilation strategies and targeted pharmacotherapy over the past three decades, pulmonary hypertension remains a significant problem for critically ill infants accounting for up to 4% of all admissions to tertiary neonatal ICUs. Reference Jain AMcNamara4
Pulmonary hypertension in infants is highly heterogeneous due to differences in pathophysiology, clinical presentation, and comorbidities. Reference Hansmann5 Despite this, pharmacological treatment has remained largely unchanged. Reference Fike and Aschner6 Recently, it has been recognised that treatment approaches should be tailored to the multiple distinct clinical phenotypes of infants with pulmonary hypertension. Reference Hilgendorff, Apitz and Bonnet7 Additionally, the emergence of novel drugs to target different mechanistic pathways has opened up new treatment options for infants with pulmonary hypertension. Reference Hansmann, Sallmon and Roehr8
Inhaled nitric oxide and phosphodiesterase inhibitors are the mainstay of contemporary neonatal pulmonary hypertension treatment. Reference Sharma and Callan EKonduri9 At the same time, endothelin receptor antagonists and prostacyclins are increasingly being investigated and used off-label. Reference Sharma and Callan EKonduri9 This has led to a rise in reports of combination therapy for high-risk infants, despite a lack of pharmacoepidemiology data to help guide treatment decisions and provide prognostic information. Reference Filan, McDougall and Shekerdemian10 Given the severity of outcomes for infants with pulmonary hypertension and the paucity of data on combination therapy, reporting real-world pharmacotherapy treatment strategies in this population is critical. This would aid in identifying the ideal combination or sequence of therapies that would benefit infants with a specific clinical phenotype.
In this study, we used a large multicenter electronic clinical data warehouse to describe the pharmacoepidemiology of critically ill infants requiring combination therapy with pulmonary vasodilators.
Materials and methods
Data source
We collected data from an electronic health record database that prospectively captures information generated by clinicians on infants cared for by the Pediatrix Medical Group in 446 neonatal ICUs in North America from 1997 to 2020. This information is gathered from routine clinical care documentation, including admission notes, daily progress notes, and discharge summaries. The record consists of data on multiple aspects of care, including demographics, maternal history, medications, procedures, laboratory results, and diagnoses. These de-identified data were transferred to the Pediatrix Clinical Data Warehouse for quality improvement and research. Reference Spitzer, Ellsbury and Handler11 This study was approved by the Duke University Institutional Review Board with a waiver of consent.
Study population
We included infants exposed to at least one of inhaled nitric oxide, intravenous or enteral sildenafil, intravenous epoprostenol, or enteral bosentan for greater than two consecutive days to ensure adequate time for simultaneous combination exposure. We excluded infants who were outborn, had missing information on discharge status, and were transferred to another institution at the time of discharge.
Definitions
We defined combination therapy as simultaneous exposure to at least two pulmonary vasodilator medications for greater than two consecutive days at any point during hospitalisation. This criterion was established to distinguish genuine combination therapy from a transition between therapies. We defined monotherapy as exposure to only one pulmonary vasodilator medication at a given time for greater than two successive days during hospitalisation. We defined pulmonary vasodilator use at discharge or death as exposure on the day of or prior to discharge or death, respectively.
We defined infants as having congenital heart disease (CHD) if they had documentation of any major (e.g. tetralogy of fallot, transposition of the great arteries, hypoplastic left heart syndrome, total anomalous pulmonary venous return) and minor heart defects (e.g. atrial septal defect, ventricular septal defect) as a clinical diagnosis in the electronic health record. We defined infants as having a patent ductus arteriosus, persistent pulmonary hypertension of the newborn, or congenital diaphragmatic hernia (CDH) if it was documented as a clinical diagnosis in the electronic health record. We defined lung anomalies as documentation of any of these clinical diagnoses in the electronic health record: cystic lung disease; lobar emphysema; pulmonary agenesis, sequestration, and lung hypoplasia; and surfactant protein abnormalities. We defined bronchopulmonary dysplasia (BPD) as continuous respiratory support (supplemental oxygen, nasal cannula, high-flow nasal cannula, nasal continuous positive airway pressure, conventional mechanical ventilation, or high-frequency ventilation) between 36 0/7 and 36 6/7 weeks postmenstrual age for infants < 32 weeks gestational age or between postnatal age 28 to 34 days for infants ≥ 32 weeks gestational age, as previously reported. Reference Trembath, Hornik and Clark12
We defined inotrope receipt as postnatal exposure to dobutamine, dopamine, epinephrine, norepinephrine, vasopressin, or milrinone. We defined diuretic receipt as postnatal exposure to acetazolamide, amiloride, bumetanide, chlorothiazide, diazoxide, ethacrynic acid, furosemide, hydrochlorothiazide, spironolactone, or metolazone. We defined receipt of postnatal steroids as exposure to dexamethasone, hydrocortisone, methylprednisolone, prednisolone, or prednisone.
We defined exposure to extracorporeal membrane oxygenation and the presence of a tracheostomy if it was documented as a procedure in the electronic health record. We defined non-invasive respiratory support as exposure to hood oxygen, nasal cannula, high-flow nasal cannula, or non-invasive positive pressure ventilation. We defined mechanical ventilation as exposure to conventional mechanical ventilation or high-frequency ventilation. We defined ventilator days as the number of days exposed to invasive mechanical ventilation (conventional or high-frequency). We defined oxygen therapy as exposure to a fraction of inspired oxygen content > 21% and oxygen days as the number of days exposed to a fraction of inspired oxygen content > 21%.
Data collection
We collected the following variables: gestational age, birth weight, small for gestational age status, race/ethnicity, sex, 5 minute Apgar score, type of delivery, inborn status, death status, antenatal steroid exposure, postnatal medications (pulmonary hypertension medications, inotropes, diuretics, postnatal steroids), diagnoses, and cardiorespiratory therapies as defined above.
For each individual and combination use of pulmonary hypertension medication(s), we collected the gestational age; postnatal age and postmenstrual age at first exposure; duration of exposure during hospitalisation; and frequency of use at any time during hospitalisation, at discharge home, and at death during hospitalisation.
Statistical analysis
We used frequencies (with percentages) and medians (with 25th and 75th percentiles) to describe categorical and continuous study variables, respectively. We compared the distribution of clinical variables between infants receiving monotherapy versus combination PH therapy using the χ2 test for categorical variables and the Wilcoxon rank-sum test for continuous variables. Additionally, we reported the following drug utilisation characteristics for each individual and combination pulmonary hypertension medication: gestational age of infants exposed to medication; postnatal age and postmenstrual age at first exposure; duration of exposure; and frequency of use at any time during hospitalisation, at discharge home, and at death during hospitalisation. We graphically represented the progression of individual and combination use over time using line graphs, stacked area graphs, and Alluvial diagrams. Reference Rosvall, Bergstrom and Rapallo13
We conducted three separate subgroup analyses on drug utilisation characteristics by focusing on infants with BPD, CHD, or CDH. Our aim was to identify any specific practice patterns with respect to pulmonary hypertension therapy within these particular subgroups.
We defined statistical significance as a p-value < 0.05. We performed all statistical analyses using Stata 17.0 (StataCorp, College Station, Texas).
Results
Infant characteristics
We identified 7681 infants that met the study inclusion and exclusion criteria (Supplementary Figure S1). These infants had a median gestational age of 31 (25th, 75th percentile: 25, 37) weeks and a median birth weight of 1557 (730, 3033) grams. Of these infants, 2883 (38%) were diagnosed with bronchopulmonary dysplasia, 1144 (15%) had CHD, 4543 (59%) had a patent ductus arteriosus, 5978 (78%) experienced persistent pulmonary hypertension of the newborn, 791 (10%) had at least one lung anomaly, 291 (4%) were born with congenital diaphragmatic hernia, and 3959 (52%) infants were exposed to prenatal steroids. At total of 5836 (76%) infants in our cohort survived to discharge.
Of the infants in our study cohort, 664 (9%) received simultaneous combination therapy at some point during their hospitalisation (Table 1). These infants were smaller and less mature compared to those receiving monotherapy, with a median gestational age of 28 (25, 37) versus 31 (26, 37) weeks (p < 0.001) and median birth weight of 1016 (590, 2760) versus 1610 (750, 3055) grams (p < 0.001), respectively; and had a higher in-hospital mortality rate [268 (40%) vs. 1577 (22%), p < 0.001].
Table 1. Infant characteristics
Variables presented as frequency (%).
PPHN = persistent pulmonary hypertension of the newborn; SGA = small for gestationalage.
* Cystic lung disease; lobar emphysema; pulmonary agenesis, sequestration, & lung hypoplasia.
Treatment or support before pulmonary hypertension medication exposure
Before an infant’s first pulmonary hypertension medication exposure, 2010 (26%) were exposed to at least one inotrope, 1253 (16%) received at least one diuretic, and 1203 (16%) received postnatal steroids. Mechanical ventilation was required in 4192 (55%) infants, 6794 (88%) needed oxygen therapy with a median fraction of inspired oxygen content of 75% (40, 100), 38 (< 1%) received a tracheostomy, and 19 (< 1%) required extracorporeal membrane oxygenation before the first pulmonary hypertension medication exposure.
Infants receiving combination therapy at any point during their hospitalization were more likelyto be exposed to diuretics and inotropes, require respiratory support with mechanical ventilation for alonger duration and higher number of oxygen days, receive a tracheostomy, and require extracorporealmembrane oxygenation before an infant’s first exposure to a pulmonary hypertension medication (Supplementary Table S2).
Pulmonary hypertension medication utilization
Among all infants in the Pediatrix database, the prevalence of exposure to any of the four pulmonary hypertension medications increased from 1997 to 2020 (Figure 1). Inhaled nitric oxide utilisation increased from 0.06% in 1997 to 0.84% in 2020; sildenafil was first used in 2003 and increased in prevalence from 0.01% to 0.16% in 2020; epoprostenol and bosentan were first used in 2005 and both grew in prevalence from < 0.01% to 0.02% in 2020. Similarly, the prevalence of monotherapy and combination therapy increased over time (Figure 1). Monotherapy increased from 0.06% in 1997 to 0.80% in 2020. We did not observe combination therapy until 2003 and utilisation increased from 0.01% to 0.10% in 2020.
Figure 1. Prevalence of pulmonary hypertension medication exposure by discharge year among all infants in the pediatrix database.
The frequency, gestational age of infants exposed, postnatal age and postmenstrual age at first exposure, duration of therapy, and use at discharge or time of death varied among each pulmonary hypertension medication and any combination of the four medications (Table 2 and Supplementary Table S2). Inhaled nitric oxide was administered to 7270 (95%) infants at some point during their hospitalization, followed by sildenafil [1177 (15%)], epoprostenol [95 (1%)], and bosentan [78 (1%)].
Table 2. Any pulmonary hypertension medication use
Continuousand categorical variables presented as median (25th, 75th percentile) andfrequency (%), respectively.
iNO = inhaled nitric oxide; PNA = postnatal age; PMA = postmenstrual age.
Inhaled nitric oxide and epoprostenol were mainly used in the very and late preterm population [median gestational age of 31 (26, 37) and 36 (29, 38) weeks, respectively] at an early median postnatal age [1 (0, 5) and 4 (1, 28) days, respectively] and for a short median duration [6 (4, 10) and 8 (5, 18) days, respectively]. On the other hand, sildenafil and bosentan were mainly used in the extreme and very preterm population [median gestational age of 27 (25, 36) and 26 (24, 32) weeks, respectively] at a much later median postnatal age [59 (13, 108) and 132 (78, 173) days, respectively] and for a longer median duration [28 (12, 66) and 31 (9, 84) days, respectively] (Table 2 and Figure 2).
Figure 2. Proportion exposed to pulmonary hypertension medications by postnatal age among all infants included our study.
Inhaled nitric oxide was mainly used as monotherapy [7104 (98%) with any single use vs. 648 (9%) with any combination use]; sildenafil was utilised as both monotherapy and combination therapy [968 (82%) vs. 631 (54%)]; and epoprostenol [18 (19%) vs. 85 (89%)] and bosentan [6 (8%) vs. 76 (97%)] were mainly used as combination therapy (Table 2). A similar pattern was seen for infants who died on pulmonary hypertension therapy [957 (86%) died on monotherapy vs. 150 (14%) died on combination therapy with inhaled nitric oxide; 119 (45%) vs. 147 (55%) with sildenafil; 4 (16%) vs. 21 (84%) with epoprostenol; and 3 (12%) vs. 23 (88%) with bosentan]. Conversely, only a small number of infants [17 (<1%)] were discharged on combination therapy and all were prescribed sildenafil and bosentan.
Dual therapy with inhaled nitric oxide and sildenafil was the most common combination therapy [599 (8%)], followed by triple therapy with inhaled nitric oxide, sildenafil, bosentan [49 (1%)] (Supplementary Table S2). The most common combination therapy that did not include inhaled nitric oxide was dual therapy with sildenafil and bosentan [33 (1%)]. Only 7 (<1%) infants were exposed to quadruple therapy. Most were extremely or very preterm infants [median gestational age of 28 (26, 38) weeks] and received this therapy at a late postnatal age [94 (53, 173) days] and for a very long median duration of 131 (44, 286) days.
There were between 0 and 6 changes to the pulmonary hypertension medication combinations throughout an infant’s neonatal ICU hospitalisation, with the initial and final medication(s) routinely differing from each other (Supplementary Figure S2).
In subgroup analysis, we found that: (1) infants with BPD received iNO at an earlier gestational age as compared to all infants in our main analysis [median gestational age of 26 (24, 30) and 31 (26, 37) weeks, respectively] (Supplementary Table S3), (2) infants with CHD received epoprostenol at a later PNA as compared to all infants in our main analysis [median PNA of 20 (4, 138) and 4 (1, 28) days, respectively] (Supplementary Table S4), and (3) all pulmonary hypertensive medications in infants with CDH were used at term or near-term median gestational ages (Supplementary Table S5). All other drug utilisation characteristics, including duration of use and frequency of use were similar between all infants in the main analysis and infants in each subgroup analysis.
Discussion
In this large multi-center observational study, we described the use of combination pulmonary hypertension drug exposure in a cohort of critically ill infants discharged from the NICU. We found that utilisation of monotherapy and combination treatment for all medications increased over time, with inhaled nitric oxide and sildenafil most used individually and concurrently. Infants exposed to combination therapy were smaller and less mature, had more cardiopulmonary disease, and were more likely to receive cardiorespiratory support than infants receiving only monotherapy. This pattern suggests that infants with greater illness severity were higher utilisers of combination therapy. The drug utilisation characteristics for each combination of medications varied substantially. Most infants experienced numerous changes in their treatment regimen, with initial and final medication combinations frequently differing. These findings suggest a need for more clarity regarding the appropriate combination of pulmonary hypertension medications, the optimal sequence of their use, and the specific clinical phenotypes for which they should be employed.
The aetiology of pulmonary hypertension in infants is multifactorial, attributable to variations in pathophysiology, clinical presentation, and comorbidities. Reference Hansmann5 It can be categorised into numerous types, including but not limited to persistent pulmonary hypertension of the newborn, bronchopulmonary dysplasia-associated pulmonary hypertension, and CHD-associated pulmonary hypertension. Reference Fike and Aschner6 Each of these clinical phenotypes can present at different postnatal ages with variable severity and disruptions in pulmonary vascular signalling pathways, which can have distinct implications for clinical management. Remodelling of the pulmonary vasculature in pulmonary hypertension can be attributed to the breakdown of one of the following signalling pathways: nitric oxide-cGMP, prostacyclin-cAMP, and endothelin receptor. Reference Nees, Rosenzweig and Cohen14 The main drugs for pulmonary hypertension aim to address perturbations in these pathways. Inhaled nitric oxide and phosphodiesterase-5 inhibitors, such as sildenafil, increase cGMP levels and cause vasodilation of the pulmonary vasculature. Prostacyclins, such as epoprostenol, initiate vasodilation by increasing cAMP levels and inhibiting pulmonary artery smooth muscle cell proliferation. Endothelin receptor antagonists, such as bosentan, augment pulmonary vasodilation by inhibiting endothelial-mediated vasoconstriction.
Despite these various therapies, only inhaled nitric oxide has gained Food and Drug Administration (FDA) approval as a pulmonary vasodilator in infants. However, this approval is limited to term and near-term infants with persistent pulmonary hypertension of the newborn. Reference Clark, Kueser and Walker15 Regardless, off-label use of inhaled nitric oxide in infants has become widespread for treating multiple phenotypes of pulmonary hypertension or preventing the development of bronchopulmonary dysplasia. Reference Hasan, Potenziano and Konduri16 Our findings affirm this use based on the following observations: (1) it was the most commonly administered therapy with 95% of infants receiving it, (2) its use increased almost 15-fold over the past two decades, (3) it was mainly used early and for a short duration, in line with its FDA approved indication, and (4) it was used in a wide range of gestational ages. These characteristics are consistent with prior multicenter epidemiologic evaluations of inhaled nitric oxide in infants. Reference Ellsworth, Harris and Carey17
We found that sildenafil was the second most frequently used pulmonary hypertension therapy, administered to 15% of infants, with a similar increase in utilisation as inhaled nitric oxide during our study. These infants were almost exclusively premature yet usually received sildenafil at near-term or term postmenstrual ages and for a median duration of nearly one month. This is consistent with current recommendations and off-label use of sildenafil as treatment for bronchopulmonary dysplasia-associated pulmonary hypertension and as adjuvant therapy to inhaled nitric oxide in premature infants with persistent pulmonary hypertension of the newborn. Reference Abman, Hansmann and Archer18,Reference Thompson, Perez and Hornik19
Similar to sildenafil, bosentan was solely used in preterm infants, at post-term postmenstrual ages, and for a median duration of approximately one month. This is despite bosentan only being FDA-approved in 2017 for treating specific types of pulmonary hypertension in children > 3 years of age. Reference Barst, Ivy and Dingemanse20 These characteristics of bosentan use in our cohort are consistent with reports and guidelines mentioning endothelin receptor antagonists as a potentially beneficial therapy for bronchopulmonary dysplasia-associated pulmonary hypertension and late or refractory persistent pulmonary hypertension of the newborn. Reference Maneenil, Talek and Thatrimontrichai21
In our study, epoprostenol had a similar drug utilisation profile to inhaled nitric oxide: early exposure, short duration of therapy, and use in near-term and term infants – characteristics congruent to infants having persistent pulmonary hypertension of the newborn. Although there is limited data about the use of prostacyclin analougs in infants, a handful of reports demonstrate clinical improvements in near-term or term infants with persistent pulmonary hypertension of the newborn. Reference Verma, Lumba and Kazmi22 However, there are also several reports showing prostacyclin use for severe bronchopulmonary dysplasia-associated pulmonary hypertension and persistent pulmonary hypertension of the newborn that is refractory to other interventions. Reference Ferdman, Rosenzweig and Zuckerman23 The characteristics exhibited by these infants do not align with those observed in our cohort receiving epoprostenol monotherapy, emphasising the importance of defining clinical phenotypes that would benefit from each pulmonary hypertension therapy.
Adult guidelines have described clinical phenotypes amenable to specific therapeutic targets at different points along the natural history of pulmonary hypertension. Reference Dweik, Rounds and Erzurum24 This has led to several studies demonstrating that combination therapy can improve clinical outcomes through additive or synergistic effects. Reference Sitbon and Gaine25 Similar phenotype-driven studies are lacking in infants and children, with only a handful of studies available on combination therapy. Reference Steffes and Austin26 This is even though combination therapy has increased 10-fold over the last two decades in our study.
Our study found that concurrent utilisation of inhaled nitric oxide with other pulmonary vasodilators was the most common in hospitalised infants, especially dual or triple therapy with enteral medications such as sildenafil and bosentan. It is important to note that these exposures typically took place at term or post-term postmenstrual ages and for relatively brief periods. This short overlap is common practice in facilitating the weaning of inhaled nitric oxide and transition to enteral medications in critically ill infants. Reference Hussain, Bondi and Shah27
We also observed that sildenafil was used frequently in combination with drugs other than inhaled nitric oxide, with the most frequent being bosentan. These infants were exclusively premature and received therapy at a later age and longer duration than treatment with sildenafil alone. All infants discharged home on combination therapy in our study received both medications. This was also recently demonstrated in a single-center randomised controlled trial of 40 infants with high-risk pulmonary hypertension receiving dual therapy with sildenafil and bosentan compared to sildenafil alone, which showed decreased pulmonary arterial pressures. Reference Vijay Kumar, Natraj Setty and Jayaranganath28 This is similarly consistent with adult studies demonstrating that dual therapy with a phosphodiesterase-5 inhibitor and an endothelin receptor antagonist is the most effective and widely utilised combination for pulmonary hypertension. Reference Dardi, Manes and Palazzini29
In our study, both bosentan and epoprostenol were primarily used as combination therapy with other pulmonary vasodilators, yet never together as dual therapy. This is even though combination therapy with prostacyclins and endothelin receptor antagonists has increased in frequency in children and adults. Reference Douwes, Zijlstra and Rosenzweig30 Additionally, we observed that any dual or triple therapy regimen involving epoprostenol occurred early during an infant’s hospitalisation and generally for less than one week. There are rare reports of intravenous prostacyclin used in combination therapy in infants with pulmonary hypertension, but all similarly involve early and brief treatment of severe persistent pulmonary hypertension of the newborn. Reference Kinugasa, Horigome and Sugiura31 Various factors may contribute to the limited use of concurrent therapy with prostacyclins in infants, including the absence of easy and non-invasive administration methods, the short duration of their effectiveness, and the severe hemodynamic side effects if not closely monitored and adjusted. Reference Fike and Aschner6
In adults, there is growing evidence from randomised controlled trials that triple therapy targeting each pathogenic pulmonary hypertension signalling pathway provides optimal disease control and improved long-term outcomes, especially in severe disease phenotypes. Reference Sitbon and Gaine25 A recent single-center retrospective observational study of 21 children with severe pulmonary hypertension who received upfront triple combination therapy with sildenafil, bosentan, and a prostacyclin found improved 3-year transplant-free survival rates of almost 90%. Reference Haarman, Lévy and Roofthooft32 Unfortunately, reports of triple therapy are limited in infants with pulmonary hypertension. Reference Radicioni, Bruni and Camerini33 Our study revealed that triple or quadruple therapy was rare, as only 1% of infants in our cohort were treated with this approach. Further, drug utilisation characteristics did not follow any discernible pattern and varied substantially depending on the combination of medications used.
The strengths of our study include (1) a large sample size, (2) a multi-center study, (3) a long study period, (4) granular drug utilisation data, and (5) the ability to follow infants’ medication regimen throughout their hospitalisation. With over 7500 infants included in our study, this is the largest observational cohort study investigating the use of combination therapy for pulmonary hypertension in critically ill infants. Furthermore, the more than 400 neonatal ICUs included in our study encompass a diverse group of academic and community facilities across North America, enhancing the universality of our results. Our study also included infants over an almost 25-year period, allowing us to track trends of combination and monotherapy use over time. Given the comprehensive information contained within the Pediatrix database, we were able to elucidate detailed drug utilisation characteristics with respect to age, frequency, timing, and duration of exposure for each combination of medications. This information helped us establish a pattern of drug usage and match it with a specific disease phenotype. Finally, given that we could obtain medication data from birth to discharge or death, this allowed us to track and visualise the complex nature of sequential or combination therapy during the natural history of pulmonary hypertension in infants.
This study has several limitations typical of observational cohort studies that rely on a large electronic health record database for administrative purposes. First, medication data is only presented at daily intervals, which precluded us from determining the exact time of day that an infant started, stopped, or transitioned to another medication. This could have overestimated the number of infants included in our study and the duration of exposure to each combination therapy. Second, there is no indication information in the medication data, so we could not ascertain whether prescribed drugs were intended for treating pulmonary hypertension. This hindered our ability to make definitive connections between select combination therapies and distinct pulmonary hypertension phenotypes. Third, our data source does not include information on the route of administration, and therefore we could not delineate differences in practice patterns between intravenous and enteral sildenafil. Fourth, our data source does not include or has limited data on several newer off-label pulmonary hypertension drugs, such as tadalafil, ambrisentan, selexipag, iloprost, and treprostinil. Fifth, we did not have access to data from echocardiograms, cardiac catheterizations, or dependable diagnostic information to precisely ascertain the presence of pulmonary hypertension. This limitation might have led to overestimating infants being treated for pulmonary hypertension, as some of these drugs could be prescribed for hypoxic respiratory failure or other pulmonary-related conditions. Sixth, our study is descriptive and does not adjust for the numerous confounders of illness severity between the combination and monotherapy groups. Therefore, any differences between the two groups should be considered descriptive and not necessarily indicative of the clinical efficacy of combination pulmonary vasodilator therapy. Lastly, we lacked detailed information on hemodynamic data and surgical procedures, including the type and timing of surgeries conducted in our CHD infants. It’s common for pulmonary hypertension medications like iNO and sildenafil to be prescribed peri-operatively for these infants. Considering that CHD infants comprised 15% of our study group, having data on whom and when these vasodilator therapies are administered would be valuable to provide further insights into the utilisation of these medications.
In conclusion, we found a significant increase and wide variation in the deployment of combination pulmonary vasodilator therapies among critically ill infants, with iNO and sildenafil constituting the most prevalent regimen. Additionally, our data found a tendency for iNO and epoprostenol use for younger, full-term infants over shorter durations compared to the use of sildenafil and bosentan. The study illuminates the urgent requirement for tailored research to define the natural progress of distinct clinical phenotypes linked to infantile pulmonary hypertension, develop strategies for identifying the ideal combination of medications for each phenotype, and elucidate the optimal timing, sequence, and dosing that would yield maximum benefits in combination therapy. The findings of our study can help inform the design and implementation of future high-quality prospective studies of combination therapy in infants.
Supplementary material
The supplementary material for this article can be found at https://doi.org/10.1017/S1047951124025976.
Acknowledgements
The authors thank the Duke Clinical Research Institute and the Pediatrix Medical Group for data acquisition and support.
Financial support
This work was funded by the Duke Clinical Research Institute’s R25 Summer Training in Academic Research (STAR) Program (grant 5R25HD076475-10; PI Daniel K. Benjamin Jr), the National Institute of Child Health and Human Development contract for the Pediatric Trials Network (grant HHSN275201000003I; PI Daniel K. Benjamin Jr), and in part by the NCATS Trial Innovation Network (grant U24TR001608; PI Daniel K. Benjamin Jr). The content is solely the authors’ responsibility and does not necessarily represent the official views of the National Institutes of Health.
Competing interests
Rachel G. Greenberg has received support from industry for research services (https://dcri.org/about-us/conflict-of-interest/). Danny K. Benjamin, Jr. reports consultancy for Allergan, Melinta Therapeutics, and Sun Pharma Advanced Research Co. Christoph P. Hornik declares consulting fees from UCB Pharma, SC Pharma, Anavex, and Lightship. Karan R. Kumar declares consulting feeds from Purdue Pharma. Jan Hau Lee’s institution receives research grant support from the National Research Medical Council, Singapore and Thrasher Foundation, United States. The other authors have nothing to disclose.
Open access
