Cardiopulmonary physiological effects of diuretic therapy in preterm infants with chronic pulmonary hypertension

Using targeted neonatal echocardiography (TNE) to examine cardiopulmonary physiological impact of diuretics in preterm infants with chronic pulmonary hypertension (cPH). Retrospective study comparing TNE indices pre- and ≤2 weeks (post) of initiating diuretic therapy in infants born <32 weeks gestational age with cPH. Twenty-seven neonates with mean gestational age, birthweight and interval between pre-post diuretic TNE of 27.0 ± 2.8 weeks, 859 ± 294 grams, and 7.8 ± 3.0 days respectively were studied. Diuretics was associated with improvement in pulmonary vascular resistance [pulmonary artery acceleration time (PAAT); 34.27(9.76) vs. 40.24(11.10)ms, p = 0.01), right ventricular (RV) ejection time:PAAT ratio [5.92(1.66) vs. 4.83(1.14), p < 0.01)], RV fractional area change [41.6(9.8) vs. 46.4(6.5%), p = 0.03)] and left ventricular myocardial performance index [0.55(0.09) vs. 0.41(0.23), p < 0.01)]. Post-treatment, frequency of bidirectional/right-to-left inter-atrial shunts decreased significantly (24% vs. 4%, p = 0.05). Primary diuretic treatment in neonates with cPH may result in improvement in PVR, RV and LV function and compliance.


INTRODUCTION
Chronic pulmonary hypertension (cPH) in infants with bronchopulmonary dysplasia (BPD) is a well-recognized pathology frequently seen in neonatal intensive care units (NICUs), associated with increased short and long-term morbidity and mortality [1].The overall incidence of cPH in infants with BPD has been reported to be between 16 and 25%, directly related with the severity of BPD.In the most severe cases, survival is reported to be as low as 47% 2 years after diagnosis [2,3].However, currently there is limited evidence on the efficacy and effectiveness of various management strategies.A recent survey of tertiary NICUs in Canada and US demonstrated significant variability in the diagnosis and management principles across sites [4].Typically, management strategies include supportive care by optimizing respiratory support, treatment of associated right ventricular (RV) congestion and specific treatment with pulmonary vasodilators, none of which are labeled for the indication.In adult patients with pulmonary hypertension, diuretic therapy is often employed to relieve symptoms and manage RV congestion [5].Recent consensus-based guidelines have also recommended diuretics as first-line pharmacological therapy in symptomatic preterm neonates with cPH [6]; however, its cardiopulmonary physiological effects are not known in this population of patients, limiting the clinical translation of these guidelines.
Since 2014, we have followed a standardized management algorithm for care of preterm neonates with cPH secondary to BPD in out tertiary NICU at Mount Sinai Hospital in Toronto, which included first-line use of diuretics for symptomatic patients.This pre-post treatment echocardiographic physiological study was conducted with the objective of evaluating the immediate cardiopulmonary physiological effects of primary treatment with diuretics in preterm infants with cPH.Based on our anecdotal experience, we hypothesized that diuretic therapy would result in improvement in echocardiographic markers of pulmonary hypertension in preterm neonates with cPH.

Study design
We conducted a single-center retrospective pre-post treatment echocardiography study from July 2014 to July 2018 at the NICU of Mount Sinai Hospital, a large tertiary perinatal unit affiliated with the University of Toronto, Canada.Our NICU primarily caters to high-risk inborn extreme premature neonates.Institutional research ethics board approval was obtained and informed consent was waived for this retrospective study.

Patient selection
For this study, all very preterm infants who had an echocardiographic diagnosis of cPH during their NICU stay secondary to BPD, defined as need for respiratory support ≥28 days post menstrual age (PMA), and received supportive treatment with diuretics were considered eligible for inclusion.In our unit, it is standard practice to use diuretics as first-line therapy for cPH when infants are deemed symptomatic.Specific inclusion criteria included gestational age (GA) at birth <32 weeks, postnatal age at cPH diagnosis on TNE and initiation of diuretics for it ≥28 days PMA, and availability of a comprehensive TNE for analysis within 24 hours prior and ≤2 weeks after starting diuretic therapy.Patients receiving other concomitant therapies for cPH at the time of pre-or post-diuretics scan or those already receiving long-term diuretic therapy for BPD were excluded as it would not be feasible to attribute changes in TNE findings to specific therapies.Further, neonates who received diuretics treatment specifically for the management of pulmonary over-circulation from intraor extra-cardiac shunts, without a diagnosis of cPH, were also not included as the purpose of this study was to evaluate treatment impact in cPH and not exclusive volume overload.The potentially eligible patients were identified from the reports in our TNE database during the study period and the health charts were subsequently reviewed to confirm eligibility.

Definition and management of cPH in the NICU
In our center we use the standard proven echocardiographic definition to diagnose cPH in preterm neonates with BPD [2], which includes presence of any of the following features: (i) flattening of interventricular septum in end-systole (seen in the short axis view at the level of the papillary muscles), (ii) right ventricular systolic pressure (RVSP), estimated from tricuspid regurgitant (TR) jet, of ≥40 mmHg and (iii) presence of a bidirectional or right-to-left shunt through a patent ductus arteriosus (PDA).A uniform cPH treatment guideline has been in use in our unit since 2014, specifically targeting symptomatic infants [7].The key components of management included (i) TNE screening for cPH for all very preterm neonates needing respiratory support between 34 and 36 weeks postmenstrual age PMA (ii) first-line treatment with diuretics for infants deemed symptomatic and have evidence of RV dilatation on TNE, (iii) frequent TNE monitoring for all neonates with cPH and pulmonary vasodilator therapy restricted for cases with progressive disease despite diuretics, and (iv) expectant management of cPH in patients without RV dilatation or if symptoms deemed mild/absent by attending clinicians, taking into context the expected course and underlying severity of BPD.Patients with cPH and signs consistent with RV congestion, such as unexplained worsening of respiratory status, excessive weight gain and peripheral edema were typically considered symptomatic, while presence and significance of other symptoms such as longer than expected need for positive pressure ventilation were adjudicated by the clinical team.In all cases, the diuretic therapy used in cPH was furosemide for 3 days (1 mg/ kg/dose twice daily if given intravenously or 2 mg/kg/dose twice daily if given enterally due to the known difference in bioavailability between both routes), followed by maintenance with aldactazide at a dose of 1 mg/ kg/day twice daily enterally, which is continued until the infant comes off respiratory support and/or TNE findings of cPH resolves.Our diuretics regime was selected based on the combination of published guidance of management of congestive heart failure in children [8] and institutional practices and preferences for long-term diuretics use in preterm infants.Our goal was to ensure decongestion of lungs and right heart with an effective dose of loop diuretics (furosemide in our center), typically adjudicated based on reduction in weight gain, followed by maintenance therapy with a potassium sparing agent (aldactazide in our center).Our unit's clinicians preferred avoiding prolonged use of frusemide in preterm neonates, when possible.Patients who worsen despite diuretics were typically considered for sildenafil treatment and/or cardiac catheterization.Typically, all patients with cPH were monitored with TNE every 2-3 weeks to guide management.

Echocardiography methods
During the study period, all TNEs were performed using an E9 ultrasound system and a 12 MHz neonatal transducer (GE Medical Systems, Milwaukee, WI, USA).The scans were performed either by a trained sonographer or TNE fellow and were reported by a staff neonatologist with expertize in TNE.For this study, all included TNEs were retrieved from our cardiology server and reanalyzed, in a random order using a dedicated workstation (EchoPAC, version 202, GE Healthcare, Horten, Norway).The following TNE indices were collected and measured using previously published methods [9].
Indices of pulmonary hemodynamics: (i) pulmonary artery acceleration time (PAAT), measured from the pulse wave Doppler obtained in the main pulmonary artery.A shorter PAAT or higher ratio adjusted to RV ejection time (RVET/PAAT) indicates higher PVR [10,11], (ii) left ventricle (LV) eccentricity index, a quantitative measure of degree of septal flattening, measured from the short axis view at the level of the papillary muscles.The ratio of the antero-inferior to septal-posterolateral linear dimensions of the LV cavity at end-systolic (EIs) and end-diastole (EId) were calculated, with higher values indicating elevated RV:LV pressure [12], and (iii) presence of bidirectional or right-to-left shunt at patent ductus arteriosus (PDA) or patent foramen ovale (PFO).The RVSP was not used to determine change in pulmonary pressures on pre-post TNEs as TR jet was not available in 63% of the scans.
Indices of right and left ventricular function.Systolic function: (i) Tricuspid annulus plane systolic excursion (TAPSE), measured using M-mode echocardiography with the line of interrogation passing through the lateral aspect of the tricuspid annulus while maintaining vertical alignment with the apex in the apical 4 chamber (A4C) view.The maximum vertical distance in systole was measured to quantify RV function, (ii) fractional area change (FAC) using planimetry method measured from the A4C view, by tracing the endocardial borders to measure RV areas at end-diastole and end-systole including the RV trabeculations within the cavity.The FAC (%) was calculated as [(RV area at end-diastole -RV area at end systole)/RV area at end-diastole] x 100, (iii) Simpson's biplane ejection fraction (EF) of the LV, using disc summation method, measured from both A4C and 2-chamber views of the LV.The LV volumes were traced in end-diastole and end-systole including the trabeculations within the cavity, and EF (%) calculated as, [(LV end-diastolic volume -LV end-systolic volume)/LV enddiastolic volume]*100, (iv) tissue Doppler imaging (TDI) derived peak systolic velocity (s'), obtained using pulse wave Doppler and a sample gate of 2 mm placed on the basal segment of ventricular free wall from the A4C view, (v) global peak systolic longitudinal strain (GLS) and strain rate analyzed using 2D speckle-tracking analysis software.For the RV, values obtained from all segments from the lateral wall (A4C view) and inferior wall (RV focused apical 3-chamber view) were averaged, while for the LV, the segmental values from apical-4, 2-and 3-chamber views were averaged for global values.The ventricular walls were manually traced at end-systole, width of the region of interest was reduced to the smallest allowed by software and automated tracking was visually inspected and readjusted, if needed before accepting the results.Strain rate is thought to be relatively less dependent on changes in loading conditions [13].In addition, ventricular outputs were measured using standard methods.For that, velocity time integral (VTI) were measured using pulse wave Doppler by placing the sample gate at the level of pulmonary or aortic valve for RV and LV output, respectively, and outflow tract diameter were measured at the same level in end-systole.Output was then calculated using the formula (ml/kg/min), [3.14 × (outflow tract radius)] 2 × VTI × heart rate]/ weight].
Diastolic and global function: The following TDI derived measures were obtained for both RV and LV -(i) early diastole velocity (e'), (ii) E/e' (ratio of peak early atrioventricular diastolic flow velocity and e' as an estimate of ventricular filling pressure), (iii) isovolumetric relaxation time (IVRT'), and (iv) myocardial performance index (MPI'), calculated as (IVRT + isovolumetric contraction time / ejection time).A lower e', greater E/e' and longer IVRT' indicates inferior diastolic function and higher MPI' indicates inferior global function.In addition, global early and late diastolic strain rates were also recorded using speckle tracking echocardiography as above.
Three consecutive beats were measured and averaged for all Doppler derived measurements.

Clinical data
Demographic and clinical data were collected from the electronic health records.Specifically, the hourly documented indices of cardiorespiratory stability (heart rate, systolic, diastolic, and mean blood pressure, respiratory rate and pulse oximetry) were averaged over the 24 h period on the day of pre-and post-treatment TNEs for comparison, while the average weight gain (mg/kg/day) was compared for the week before and after treatment initiation.All participants ventilatory details [type of respiratory support, mean airway pressure (MAP), fraction of inspired oxygen (FiO 2 ) and respiratory severity score (RSS = MAP × FiO 2 )], total fluid intake (ml/kg/day) and urine output (ml/kg/hour) were extracted on the day of diuretics initiation and a week after.The documented MAP and FiO 2 were averaged for the 24 h period.All infants had their serum sodium levels checked within the week after treatment onset.These data were reviewed and incidence of new onset hyponatremia (serum sodium < 135 mmol/L) were recorded and classified as mild (serum sodium 130-134 mmol/L), moderate (serum sodium 125-129 mmol/L) and severe (serum sodium < 125 mmol/L).Routine renal ultrasound to screen for nephrocalcinosis was not practiced in our unit; however, all infants' imaging results after diuretics initiation were reviewed to identify any case of a new diagnosis.Further, subsequent clinical course relevant to cPH management and discharge details were also extracted.

Outcomes
The pulmonary artery acceleration time (PAAT) on post-treatment TNE in comparison to pre-treatment was considered the primary outcome of interest, as it is well known to correlate with changes in PVR [10].Other TNE indices of pulmonary hemodynamics, RV and LV function as well as clinical indices of cardiorespiratory stability and ventilatory needs, were compared as secondary outcomes.Further, potential attributable side effects (hyponatremia and its severity, nephrocalcinosis) and subsequent relevant clinical course details (survival to discharge, discharge on home oxygen, cPH resolution pre-discharge, need for sildenafil treatment and discharge on diuretics) were also described as secondary outcomes.

Statistical analysis
Study data were examined for normality using visual inspection of histograms and Shapiro-Wilk test, and presented as number (percentage), mean (±standard deviation, SD) or median (interquartile range, IQR), as appropriate.The study population was summarized descriptively.Pre-and post-diuretic PAAT, other continuous TNE measures of pulmonary hemodynamics, RV and LV function and clinical physiological indices were compared using paired t-test.The frequency of categorical variables prepost treatment were analyzed using Chi-squared test or Fisher's exact test, as appropriate.Further, to examine the effect of PFO and the observed cardiopulmonary physiological effects of diuretics in cPH patients, a subgroup analysis was performed comparing pre-post TNE indices restricted to patients with PFO diameter less than the median for the whole cohort.A p value less 0.05 was considered statistically significant and analysis was not adjusted for multiple comparisons, given the observational nature of the study design.Analysis was conducted using IBM SPSS statistics 21.

RESULTS
In our unit, during the study period, 38 neonates received diuretic therapy specifically as first-line management of cPH.Of these, 27 fulfilled the eligibility criteria and were included for pre-post treatment TNE comparison; post-treatment TNE was not available within 14 days for 9 patients while 2 infants were already receiving low dose diuretics for BPD treatment at the time of initial TNE.The baseline characteristics of included patients are described in Table 1.The pre-treatment TNE was performed at a mean postmenstrual age of 36.5 ± 3.4 weeks and the interval between pre-and post-treatment TNE was 7.8 ± 3.0 days.

Post-treatment TNE and physiological indices
Diuretics treatment was associated with significant improvement in TNE indices reflective of PVR [increase in PAAT and reduction in RVET:PAAT ratio], as well as decrease in the frequency of right-toleft PFO shunt (Table 2).Except for slightly lower diastolic blood pressure post-treatment, the clinical cardiopulmonary physiological parameters averaged on the day of TNEs did not differ between the two time points.The vast majority of indices of RV and LV function remained unchanged on post-treatment TNE, except post-diuretics scans demonstrating a small increase in FAC and significant lowering of LV-MPI, suggesting improvement in RV systolic function and LV global function (Table 3).

Immediate clinical course
As expected, the week following initiation of diuretics demonstrated less weight gain compared to the preceding week (10.3 ± 11.3 vs. 21.9 ± 13.2 g/kg/day, p < 0.01).Although the overall number of infants needing positive pressure ventilation support was similar after a week of diuretic treatment (78% vs. 74%, p = 0.75), changes in respiratory status were observed within cohort.Fifteen (56%) patients had a reduction in respiratory support following a week of treatment (10 had lowering of RSS on same ventilatory mode, 4 came off respiratory support to room air and 1 de-escalated form NIPPV to nCPAP).Five (19%) patients, however, experienced an increase in respiratory support (2 increased in RSS, 3 escalated in mode of ventilatory support), while 7 (26%) patients' respiratory support remained unchanged after 1 week.

Subsequent clinical course
Overall, all but one infant, who died on account of severe chronic brain injury, survived until hospital discharge; 10 (37%) were discharged on home oxygen.With regards to the cPH course, 7 (26%) patients had resolution of all TNE signs suggestive of pulmonary vascular disease and diuretics were successfully discontinued before discharge, 3 (11%) infants experienced worsening despite diuretics requiring additional sildenafil treatment (1 died, 2 discharged on sildenafil and diuretics), while in 16 (59%) cases pulmonary vascular disease signs remained stable but unresolved on TNE and were discharged home on diuretics with pediatric cardiology follow up.In terms of side effects, hyponatremia occurred in 17 (63%) infants; 15 mild and 2 moderate, all of which resolved with oral sodium supplementation.There was one case of a new diagnosis of nephrocalcinosis on renal ultrasound which occurred in the context of high serum vitamin D levels.

DISCUSSION
In this physiological pre-post treatment TNE study including preterm neonates with echocardiography-proven cPH, we found that systematic early first-line treatment with diuretics resulted in significant improvement in echocardiographic markers of PVR, RV systolic and LV global performance, which was evident within 2 weeks of therapy.These changes were associated with early improvements and without subsequent deterioration in the clinical respiratory status during hospital stay for most patients.Only 3 out of 27 treated patients demonstrated further progression of cPH despite diuretics, requiring specific pulmonary vasodilator therapy.Chronic pulmonary hypertension is an important well-recognized complication of premature birth, occurring most commonly in the context of BPD and its severity [14][15][16][17].Physiologically, it is characterized by sustained elevated PVR, exposing the right ventricle to high afterload which may persist beyond infancy and into childhood [18].While it is well established that BPD complicated with cPH results in higher mortality and morbidities in preterm neonates, both in patient and after discharge, in comparison to BPD without cPH, data on its treatment remains highly sparse [16,19].While a few case series have indicated a potential role of sildenafil, a selective Type 5 phosphodiesterase inhibitor, for preterm neonates with cPH, the response to treatment was evaluated over several months and in the context of its use with several other concomitant therapies [20][21][22][23][24].The mortality rates in these patients were also high, likely a reflection of case selection where the most severe and resistant cases were given treatment; however, several patients experienced side effects such as hypotension [25].While sildenafil is the most frequently reported specific pulmonary vasodilatory agent employed in this patient population, pending further data on most effective dose, its safety and effectiveness, current clinical use is off-label and restricted to patients unresponsive to supportive care and treatments [26].
Recent European Pediatric Pulmonary Vascular Disease Network (EPPVDN) specifically recommended diuretics as the first line pharmacotherapy for preterm neonates with cPH from BPD [27].In line, at our center, we have been using a standard management algorithm since 2014, based on routine TNE screening of preterm neonates on respiratory support at 34-36 weeks PMA and first-line use of diuretics in cases of symptomatic cPH, failing which sildenafil treatment is considered.In a previous clinical outcome study at our center, including a cohort of extreme preterm neonates cared for using this approach, we found that although patients with BPD and cPH had a prolonged in-hospital respiratory course and need for home oxygen, the mortality rate and longterm respiratory and neurodevelopmental outcomes may be similar to patients with BPD without cPH [28].This observational study is an extension of our previous work examining the clinical impact of diuretics in cPH preterm patients and provides physiological data in support of the EPPVDN guidelines.
There are a few potential mechanisms which may explain our physiological echocardiographic findings.Occurrence of pulmonary edema and RV congestion is known to be central to the pathophysiology and clinical manifestations of chronically elevated PVR [29].Persistently elevated capillary hydrodynamic lateral pressure, a term used to describe the pressure across the vessel walls, leads to increased fluid leak into the interstitium and subsequent increased interstitial pressure.In return, external pressure exerted onto the capillaries, disrupt the integrity of the capillary architecture, and the re-distribution of blood in a smaller cross-sectional area causing further increased PVR.In theory, by alleviating the underlying pulmonary edema, this can lead to symptomatic relief and ultimately improvement in pulmonary Total fluid intake (ml/kg/day) 155 ( 12) 152 ( 13) 3 (−3.83,9.83) Results presented as mean (standard deviation; SD), n (%) and mean difference (95% confidence interval; CI).PFO patent foramen ovale, PDA patent ductus arteriosus, TNE targeted neonatal echocardiography.a Denotes p < 0.05, b denotes p < 0.01.Eccentricity index was calculated at end-diastole and end-systole using the D2/D1 formula where D1 is the septalposterolateral diameter and D2 the antero-inferior dimension of the left ventricle in the parasternal short axis view at the level of the papillary muscle.
Respiratory severity score was calculated using mean airway pressure × fraction of inspired oxygen equation in infants receiving either invasive or non-invasive positive pressure support (n = 21 at pre-diuretics and n = 20 post diuretics).All clinical physiological variables averaged for the 24 h period on the day of TNE.
pressures.Hence, it is plausible that the observed improvement in PVR parameters primarily represent the resolution of the contribution of interstitial edema on cPH, breaking this vicious cycle and improving pulmonary mechanics.Recent data has also indicated that inferior LV function, in particular diastolic function, may contribute to the pathophysiology of cPH in BPD, through chronic elevation of LV end diastolic pressure and subsequently left atrial and pulmonary venous pressure [30,31].By reducing systemic afterload, diuretic therapy may also improve LV end diastolic pressure and its contributory effects on pulmonary venous pressure in cPH pathophysiology and pulmonary edema.
Our findings of reduction in DBP and significant improvement in LV MPI, a marker of global function, in association with improvement in PVR markers after treatment lend support to this hypothesis.It is possible that the impact of diuretics is a function of both mechanisms, although their relative contribution is unknown and may vary between patients.Furosemide is also thought to have some direct pulmonary vasodilator effect; however, it is less likely to have an impact on our findings given the short duration of exposure (3 days) following which we primarily used aldactazide.This was chosen to minimize the potential of side-effects.Furosemide is known to be associated with several adverse outcomes such as dyselectrolytemia and nephrocalcinosis, most of which are a direct function of duration of exposure [32][33][34].Diuretic treatment is often required in critical care units to manage symptoms of systemic or pulmonary congestion.Loop diuretics have been extensively studied in preterm infants in the context of parenchymal lung disease.While its use has demonstrated improvement in lung compliance, reduced respiratory support requirements and facilitation of extubation, data from large trials have not demonstrated its on BPD prevention or a reduction in the total duration of oxygen therapy [32,35].Hence, routine use in all patients with BPD is not recommended [35].Further, as with all medication, the potential side effects of diuretic therapy will require a risk-benefit evaluation to rationalize its use.The data from our study may help rationalize the role of diuretics in preterm neonates with BPD and support future trials designed specifically for the management of patients with the additional burden of cPH.Nevertheless, our study has demonstrated that a significant number of patients develop dyselectrolytemia, specifically hyponatremia.While all cases resolved promptly with oral supplementation, it highlights the importance of monitoring of electrolytes in preterm neonates undergoing diuretic therapy.
In addition to the typical constraints of a retrospective study, the results of our study will need to be interpreted in the context of some important limitations.As this is an observational pre-post treatment study, our findings are limited by a lack of control group.While it is not expected to observe a spontaneous improvement in PVR markers within 2 weeks in patients with cPH, we cannot rule out the effect of time vs. treatment.In our study cohort, only one infant was receiving mechanical ventilation at the time of treatment initiation.Our unit's population represents a typical modern high-volume perinatal center in North America with adoption of care practices aimed at lung protection such as avoidance of mechanical ventilation and early extubation, wherever possible.Further, we have been following the practice of early diagnosis, close monitoring, and treatment of cPH in at-risk neonates since 2014, which may also have modulated disease severity in our patients over time.Hence, the incidence of the most severe form of BPD and related vascular disease may be lower in the study cohort than other referral or surgical centers.Nevertheless, this study was a physiological investigation which provides proof-of-concept data in line with the recommendations by some international expert panels.It is certainly possible at in centers with higher burden of severe cPH, diuretics therapy alone may not be as effective as in our study cohort.Additionally, we were not able to analyze all eligible patients as the post treatment TNE was not available for analysis in 9 patients.This was on account of operational changes made to our institution's echocardiographic archive system during the study period which resulted in the loss of data for some patients.Lastly, we are unable to comment on comparative efficacy with other potential management strategies including first line use of pulmonary vasodilator agents.Nevertheless, this physiological interrogation using pre-post TNEs provide the first data demonstrating potentially beneficial immediate cardiopulmonary effects of systematic diuretic treatment, lending support to future therapeutic trials to establish clinical effectiveness.CONCLUSION Among preterm neonates with cPH secondary to BPD, systematic first line use of diuretic therapy results in significant immediate improvement in echocardiographic markers of PVR in association with significant improvement in RV systolic and LV global performance.Our data suggest that it may be reasonable to consider diuretics therapy as first-line approach in the management of cPH and RV congestion in preterm neonates and support designing a large, randomized control trial to test their effectiveness in modulating clinical outcomes.

Table 1 .
Baseline characteristics of infants who received diuretic therapy for chronic pulmonary hypertension associated with bronchopulmonary dysplasia (n = 27).
Results presented as mean standard deviation (SD), median (interquartile range) or number (percentage) as appropriate; n/N to identify those variables with missing data.PROM Premature rupture of membranes; MgSO 4 magnesium sulfate; NICU neonatal intensive care unit; IVH intraventricular hemorrhage > grade 3 as per Papille's classification.PDA patent ductus arteriosus, cPH chronic pulmonary hypertension, NIPPV non-invasive positive pressure ventilation, CPAP continuous positive airway pressure.

Table 2 .
Comparison of echocardiographic indices of pulmonary hemodynamics and clinical physiological variables before and after diuretic therapy (n = 27).

Table 3 .
Comparison of echocardiographic indices of right (RV) and left ventricular (LV) dimensions, systolic, diastolic, global functions and outputs before and after diuretic therapy (n = 27).< 0.05, b denotes p < 0.01.s' is peak systolic velocity measured using tissue Doppler imaging at the basal segment of the free wall of the left ventricle from apical 4-chamber view.Longitudinal strain and strain rate were measured using speckle tracking echocardiography.The values obtained from all segments of the LV wall in apical 4-chamber, 3-chamber and 2-chamber views were averaged to give LV global strain and strain rate.E (early) and A (late) diastolic velocities of the ventricular inflow were measured using pulse wave Doppler.E/A and E/e' are markers of ventricular diastolic performance.Myocardial performance index (marker of global performance) was obtained using tissue Doppler imaging using the formula, MPI = [(isovolumetric relaxation time + isovolumetric contraction time)/right ventricular ejection time].
a Denotes p