Hydralazine in infants with persistent hypoxemic respiratory failure

  • Review
  • Intervention

Authors


Abstract

Background

Most deaths of infants with chronic lung disease (CLD) are caused by respiratory failure, unremitting pulmonary artery hypertension (PAH) with cor pulmonale, or infection. Although the exact prevalence of PAH in infants with CLD is unknown, infants with CLD and severe PAH have a high mortality rate. Except for oxygen supplementation, no specific interventions have been established as effective in the treatment for PAH in premature infants with CLD. Little has been proven regarding the clinical efficacy of vasodilators and concerns remain regarding adverse effects.

Objectives

To review current evidence for the benefits and harms of hydralazine therapy to infants with persistent hypoxemic respiratory failure.

Search methods

We searched the Cochrane Central Register of Controlled Trials (CENTRAL, The Cochrane Library), MEDLINE via PubMed and EMBASE, and other clinical trials registries through November 2011 using the standard search strategy of the Cochrane Neonatal Review Group. We searched these databases using a strategy combining a variation of the Cochrane highly sensitive search strategy for identifying randomised trials in MEDLINE; sensitivity-maximising version with selected MeSH and free-text terms: hydralazine, vasodilator agent, antihypertensive agent, heart diseases, lung diseases, respiratory tract diseases, infant, and randomised controlled trial.

Selection criteria

We considered only randomised controlled trials and quasi-randomised trials for inclusion. We included low birth weight (LBW) infants with persistent hypoxemic respiratory failure who were treated with any type of hydralazine therapy.

Data collection and analysis

Two review authors independently assessed trial quality according to pre-specified criteria.

Main results

We found no studies meeting the criteria for inclusion in this review.

Authors' conclusions

There was insufficient evidence to determine the safety and efficacy of hydralazine in LBW infants with persistent hypoxemic respiratory failure. Since hydralazine is inexpensive and potentially beneficial, randomised controlled trials are recommended. Such trials are particularly needed in settings where other medications such as sildenafil, inhaled nitric oxide (iNO), or extracorporeal membrane oxygenation (ECMO) are not available.

Résumé scientifique

Hydralazine pour les nourrissons souffrant d'insuffisance respiratoire hypoxémique persistante

Contexte

La plupart des décès chez les nourrissons atteints d'une maladie pulmonaire chronique (MPC) sont causés par une insuffisance respiratoire, une hypertension artérielle pulmonaire (HAP) chronique associée à un cœur pulmonaire, ou une infection. Bien que la prévalence exacte de l'HAP chez les nourrissons atteints d'une MPC soit inconnue, les nourrissons atteints d'une MPC et d'une HAP sévère présentent un taux de mortalité élevé. A l'exception de la supplémentation en oxygène, aucune intervention spécifique n'a été établie comme étant efficace dans le traitement de l'HAP chez les nouveau-nés prématurés atteints d'une MPC. Il existe peu de preuves concernant l'efficacité clinique des vasodilatateurs, et des interrogations subsistent concernant les effets indésirables.

Objectifs

Passer en revue les données actuelles sur les bénéfices et les préjudices du traitement par hydralazine chez les nourrissons souffrant d'insuffisance respiratoire hypoxémique persistante.

Stratégie de recherche documentaire

Nous avons effectué une recherche dans le registre Cochrane des essais contrôlés – Cochrane Central Register of Controlled Trials (CENTRAL, The Cochrane Library), MEDLINE via PubMed et EMBASE, et dans d'autres registres d'essais cliniques jusqu'à novembre 2011 en utilisant la stratégie de recherche standard du Groupe thématique Cochrane sur la néonatologie. Nous avons effectué des recherches dans ces bases de données en appliquant une stratégie combinant une variation de la stratégie de recherche très sensible de Cochrane pour identifier les essais randomisés dans MEDLINE ; version permettant de maximiser la sensibilité avec termes MeSH et sous forme de texte libre sélectionnés : hydralazine, vasodilator agent, antihypertensive agent, heart diseases, lung diseases, respiratory tract diseases, infant, and randomised controlled trial.

Critères de sélection

Nous n'avons inclus que des essais contrôlés randomisés et des essais quasi-randomisés. Nous avons inclus les nouveau-nés de faible poids de naissance (FPN) souffrant d'insuffisance respiratoire hypoxémique persistante qui étaient traités par tout type de traitement par hydralazine.

Recueil et analyse des données

Deux auteurs de la revue ont évalué la qualité des essais de manière indépendante conformément aux critères pré-spécifiés.

Résultats principaux

Nous n'avons pas trouvé d'étude remplissant les critères d’inclusion dans cette revue.

Conclusions des auteurs

Les preuves étaient insuffisantes pour déterminer l'efficacité et l'innocuité de l'hydralazine chez les nourrissons de FPN souffrant d'insuffisance respiratoire hypoxémique persistante. Puisque l'hydralazine est peu coûteuse et potentiellement bénéfique, des essais contrôlés randomisés sont recommandés. De tels essais sont particulièrement nécessaires dans le cas où les autres médications telles que le sildénafil, le monoxyde d'azote inhalé (MAi) ou l'oxygénation extracorporelle par oxygénateur à membrane (OECM) ne sont pas disponibles.

Plain language summary

Hydralazine for pulmonary hypertension in low birth weight infants with chronic lung disease

In premature infants, pulmonary arterial hypertension (PAH) associated with chronic lung disease (CLD) is associated with high mortality rate. With the exception of oxygen supplementation, no specific interventions have been established as an effective treatment for PAH in premature infants with CLD. Vasodilators could be effective treatments to reduce pulmonary arterial pressure, but little has been proven regarding their clinical effectiveness and concern remains regarding adverse effects. This review found no trials of the use of hydralazine for low birth weight infant with PAH related to CLD. However, since hydralazine is inexpensive and potentially beneficial, randomised controlled trials are recommended.

Résumé simplifié

Hydralazine pour traiter l'hypertension pulmonaire chez les nourrissons de faible poids de naissance atteints d'une maladie pulmonaire chronique

Chez les nouveau-nés prématurés, l'hypertension artérielle pulmonaire (HAP) associée à une maladie pulmonaire chronique (MPC) est associée à un taux de mortalité élevé. A l'exception de la supplémentation en oxygène, aucune intervention spécifique n'a été établie en tant que traitement efficace pour l'HAP chez les nouveau-nés prématurés atteints d'une MPC. Il se peut que les vasodilatateurs constituent un traitement efficace pour réduire la pression artérielle pulmonaire, mais il existe peu de preuves concernant leur efficacité clinique, et des interrogations subsistent concernant les effets indésirables. Cette revue n’a trouvé aucun essai sur l'utilisation de l'hydralazine chez les nourrissons de faible poids de naissance atteints d'une HAP associée à une MPC. Cependant, puisque l'hydralazine est peu coûteuse et potentiellement bénéfique, des essais contrôlés randomisés sont recommandés.

Notes de traduction

Traduit par: French Cochrane Centre 1st March, 2013
Traduction financée par: Pour la France : Ministère de la Santé. Pour le Canada : Instituts de recherche en santé du Canada, ministère de la Santé du Québec, Fonds de recherche de Québec-Santé et Institut national d'excellence en santé et en services sociaux.

Background

Description of the condition

General definition of bronchopulmonary dysplasia and chronic lung disease

In 1967, Northway et al first described bronchopulmonary dysplasia (BPD), a new pulmonary disorder that developed in premature infants exposed to mechanical ventilation and high oxygen supplementation (Northway 1967). In 1988, Shennan and co-workers demonstrated that oxygen dependency at 34 to 36 weeks' postmenstrual age (PMA) predicted worse outcome in premature infants than oxygen dependency at 28 days (Shennan 1988). In 2001, a National Institute of Child Health and Human Development/National Heart, Lung, and Blood Institute/Office of Rare Diseases workshop developed a definition of BPD that has been accepted in the clinical field (Jobe 2001; Bancalari 2006). They defined BPD as the need for supplemental oxygen for at least 28 days after birth. As less-mature infants were routinely supported in neonatal intensive care, the deficiencies of using a definition of BPD at 28 days became apparent.

Relatively more mature infants can develop BPD. Although the path to BPD or chronic lung disease (CLD) is most often due to prematurity and respiratory distress syndrome, several other conditions, such as pneumonia, sepsis, aspiration syndromes, pulmonary hypoplasia, diaphragmatic hernia, and congenital heart disease, can be a cause of CLD. The inciting factors are not only the underlying disorder, but also the effects of the supportive treatment, including mechanical ventilation, barotrauma, and oxygen toxicity (Allen 2003). For the purpose of this review, we have defined CLD as oxygen requirement at 36 weeks' PMA.

Chronic lung disease and pulmonary hypertension

Most deaths of infants with CLD are caused by respiratory failure, unremitting pulmonary artery hypertension (PAH) with cor pulmonale, or infection. PAH in infants with CLD results from a combination of factors including an absolute reduction in the size and complexity of the pulmonary vascular bed, increased resting tone of pulmonary artery smooth muscle, and increased reactivity of the arteries to a variety of stimuli (Tomashefski 1984; Bush 1990; Hislop 1990; Stenmark 2005). Although the exact prevalence of PAH in infants with CLD is unknown, infants with CLD and severe PAH have a high mortality rate (Khemani 2007; An 2010). In a study of infants with BPD treated during the recent surfactant era, those infants who developed PAH had an estimated survival rate of 64% (± 8%) at six months and 53% (± 11%) at two years after diagnosis of PAH. In multivariate analyses, small birth weight for gestational age and severe PAH (defined as systemic or supra-systemic right ventricular pressure) were associated with worse survival rates (Khemani 2007; Walter 2009).

Pulmonary circulation in patients with BPD is abnormally responsive to oxygen and other pulmonary vasodilators (Abman 1985; Mourani 2004; Stenmark 2005). Despite limited knowledge regarding the risks and benefits, long-term supplemental oxygen therapy is considered the standard treatment for PAH associated with BPD as it could decrease pulmonary vascular resistance (PVR) and thereby decrease the risk of progression to cor pulmonale (Halliday 1980; Abman 1985; Benatar 1995; STOP-ROP 2000).

Multiple other treatment strategies for PAH, including vasodilators such as hydralazine, calcium channel blockers, tolazoline, endothelin antagonists, prostacyclin, phosphodiesterase (PDE) inhibitors, and inhaled nitric oxide (iNO) have been evaluated (Greenough 2005; Ostrea 2006; Oishi 2011). Nitric oxide (NO) is one of the most promising. It acts as a vasodilator by relaxing the vascular smooth muscle cells by increasing cGMP (cyclic guanosine monophosphate) level. The long-term benefits of iNO are still unclear (Bush 1990; Mourani 2004). There are several adverse effects that need to be considered, such as methaemoglobinaemia (Bizzarro 2005). Tolazoline, one of the former frequently used treatment options, is an α-adrenergic agent and dilates vessels non-specifically. Tolazoline has been used less often because of its now well-known adverse effects, such as gastric bleeding, systemic hypotension, and oliguria. Other vasodilators mentioned above could also be effective treatments to reduce pulmonary arterial pressure, but little has been confirmed regarding their clinical effectiveness, and concern remains regarding adverse effects such as systemic hypotension (Nuntnarumit 2002; Greenough 2005; Ostrea 2006).

Description of the intervention

Hydralazine is a vasodilator used to treat patients with severe hypertension, pre-eclampsia/eclampsia, or chronic heart failure (Duley 2006; Marik 2007; Hunt 2009). Although many newer drugs have been developed for the treatment of hypertension, hydralazine is still widely used in emergency and critical care fields due to its lower cost and extensive clinical experience (Fivush 1997). The usual dose range is 0.1 to 0.2 mg/kg/dose (not to exceed 20 mg) and duration is every four to six hours as needed, up to 1.7 to 3.5 mg/kg/day divided into four to six doses for paediatric patients. The possible route of administration is oral, intramuscular, and intravenous. Known major adverse reactions are heart failure, hypotension, reflex tachycardia, neurological changes, immunological reactions such as drug-induced lupus syndrome, serum sickness, haemolytic anaemia, vasculitis, and rapidly progressive glomerulonephritis (William 2007; Kandler 2011).

How the intervention might work

Hydralazine is thought to reduce peripheral resistance directly by relaxing the smooth muscle cell layer in arterial vessels. Hydralazine does not dilate venous capacitance vessels (McGoon 1983; Zuppa 2008). The mechanism of action has not been identified, but altered Ca2+ balance in vascular smooth cells, whereby inhibition of Ca2+ release from the sarcoplasmic reticulum prevents contraction mediated by Ca2+-dependent ATPases, kinases, or ion channels (Knowles 2004), may contribute to the effect of hydralazine. Clinically, hydralazine has been used to treat right heart failure caused by pulmonary arterial hypertension. When PVR is elevated, vasodilator therapy helps the failing right ventricle by decreasing afterload. A reduced afterload may also allow a decline in right ventricular end diastolic volume (RVEDV), producing decreased wall tension and myocardial oxygen requirement. The dilated pulmonary vasculature also increases left ventricular preload, which increases mean arterial pressure and right coronary arterial (RCA) perfusion pressure (Subhedar 2000; Slesnick 2006; Mourani 2008).

Why it is important to do this review

No specific interventions have been established as a widely accepted effective treatment for PAH in premature infants with evolving CLD, except oxygen supplementation. Although iNO is achieving the status of primary treatment for PAH in infants (Barrington 2010), hydralazine may have some advantages over iNO, including extremely low cost, a variety of routes of administration, and no possibility of harm for medical staff from passive inhalation.

We planned a review of the current evidence for the benefits and harms of hydralazine therapy in infants with CLD.

Objectives

Primary

To determine the efficacy and safety of hydralazine compared to placebo or other treatments in infants with persistent hypoxemic respiratory failure.

We also planned to analyse the following subgroups:

  1. preterm (< 37 weeks' gestation) versus term infants;

  2. gestational age (< 32 weeks versus ≥ 32 weeks);

  3. extremely (< 1000 grams at birth); very low birth weight (< 1500 grams at birth); low birth weight (LBW) infants (< 2500 grams);

  4. severity of BPD (each level of BPD; using the definition of the National Institute of Child Health and Human Development/National Heart, Lung, and Blood Institute/Office of Rare Diseases Workshop 2001; Jobe 2001; Bancalari 2006);

  5. confirmed PAH prior to study entry versus unconfirmed PAH;

  6. duration of treatment with hydralazine (< 7 days versus ≥ 7 days);

  7. route of administration (oral, intramuscular, and intravenous);

  8. dose of treatment with hydralazine (< 2 mg/kg/day versus ≥ 2 mg/kg/day).

Methods

Criteria for considering studies for this review

Types of studies

We considered randomised controlled trials (RCTs) (including cluster-randomised trials) and quasi-randomised trials for this review.

Types of participants

Infants with persistent hypoxemic failure

Persistent hypoxemic failure was defined as persistent need for supplemental oxygen and assisted ventilation at greater than one week of age for any given causes except known congenital cardiac anomaly. We included all the infants who received the hydralazine treatment, whether or not they had confirmed PAH.

Types of interventions

The intervention of interest was any type of hydralazine therapy, including oral administration.

We considered studies comparing the following interventions:

  1. hydralazine compared with placebo or no treatment;

  2. hydralazine compared with other potential treatments for pulmonary hypertension with CLD: calcium channel blockers, tolazoline, endothelin antagonists, prostacyclin, PDE inhibitors, and iNO.

We planned to include any dose and duration of hydralazine therapy. The comparison interventions could be either single interventions or combination of therapies or any combination of therapies for PAH (e.g. hydralazine plus calcium blocker versus prostacyclin).

Types of outcome measures

Primary outcomes
  1. Survival at 36 weeks' PMA, in-hospital survival at hospital discharge, and at 18 and 36 months of age.

Secondary outcomes
  1. Improvement rate of PAH compared before with any timing after the intervention; improvement of PAH is defined as a tricuspid regurgitation (TR) ≤ 2.5 m/s, or a diminished amount of TR, restoration of interventricular septal configuration, regressed right ventricular hypertrophy (RVH) and dilation if using echocardiography, and pulmonary arterial pressure < 25 mmHg if assessed by cardiac catheterization.

  2. Neurodevelopment (assessed by Bayley, Griffith, or any other validated tools) assessed at adjusted age of 18 months (Black 1999).

  3. Length of hospitalisation (days) after the birth.

  4. Length of ventilation (days) after the birth.

  5. Length of oxygen supplementation (days) after the birth.

  6. Level of oxygen supplementation (FiO2 or some other measure), or measures of oxygenation (oxygenation index, arterial/alveolar oxygen ratio).

  7. Adverse events, such as heart failure, hypotension, reflex tachycardia, neurological changes, immunological reactions such as drug-induced lupus syndrome, serum sickness, haemolytic anaemia, vasculitis, and rapidly progressive glomerulonephritis (or other adverse effects based on reports in the literature).

We defined PAH using either echocardiography or cardiac catheterization. Using echocardiography, we defined PAH as one or both of the following criteria:

  1. maximal velocity of the TR jet (≥ 3 m/sec); or

  2. flat or left-deviated interventricular septal configuration, and RVH with chamber dilation (Badesch 2009).

If using cardiac catheterization, PAH was defined as pulmonary arterial pressure > 25 to 30 mmHg (Wessel 1997; Adatia 2002; Subhedar 2004).

Search methods for identification of studies

Electronic searches

We used the standard search strategy of the Cochrane Neonatal Review Group as outlined in The Cochrane Library. We searched the Cochrane Central Register of Controlled Trials (CENTRAL, The Cochrane Library), MEDLINE via PubMed and EMBASE (1966 to November 2011), and other clinical trials web sites. We also searched these databases using a strategy combining a variation of the Cochrane highly sensitive search strategy for identifying RCTs in MEDLINE; sensitivity-maximising version (Higgins 2011) with selected MeSH and free-text terms: hydralazine, vasodilator agent, antihypertensive agent, heart diseases, lung diseases, respiratory tract diseases, infant, and randomised controlled trial (publication type). The MEDLINE search strategy translated into the other databases using the appropriate controlled vocabulary as applicable. We did not apply any language restriction. We limited the search to humans and clinical trials. We did a lateral search using the 'related articles' link in PubMed for the articles initially retrieved from the search strategy.

We also reviewed the reference lists of identified articles and handsearch reviews, bibliographies of books, and abstracts. We cross-checked references from identified studies for possible additional studies. We contacted the original manufacturer of hydralazine (Novartis) to identify any additional unpublished or ongoing trials. We also searched the web sites that had registries of ongoing or recently completed trials on this subject.

Data collection and analysis

We followed the methodology for data collection and analysis in the Cochrane Handbook of Systematic Reviews of Interventions (Higgins 2011).

Selection of studies

Two review authors (Atsushi Kawaguchi and Tetsuya Isayama) independently assessed the eligibility of the trials. We selected studies as being potentially relevant by screening the titles and abstracts. We obtained the full text of the article for review when a decision could not be made by screening the title and the abstract. The two review authors retrieved the full texts of all potentially relevant articles and independently assessed the eligibility by filling out eligibility forms designed in accordance with the specified inclusion criteria. We made efforts to contact the original investigators for additional data and information when required.

Data extraction and management

We planned to extract the data using a data extraction form that was designed by the review authors. The review authors planned to extract the data independently. We made efforts to contact study investigators for additional information or data. We planned to enter data into Review Manager Software (RevMan 5.1) (RevMan 2011).

Assessment of risk of bias in included studies

We planned to use the standard methods of The Cochrane Collaboration and its Neonatal Review Group (neonatal.cochrane.org/en/index.html) to assess the methodological quality of included studies. In addition, we planned to assess study quality and risk of bias using the following criteria documented in the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2011). We also planned to assess eligible studies using the following key criteria: allocation concealment (blinding of randomisation), blinding of intervention, completeness of follow-up, and blinding of outcome measurement, though there was no trial eligible to be included in this review.

We planned to use the 'Risk of bias' table, which addressed the following questions.

(1) Sequence generation (checking for possible selection bias)

Was the allocation sequence adequately generated? 

For each included study, we planned to categorise the method used to generate the allocation sequence as: 

  • low risk (any truly random process, e.g. random number table; computer random number generator); 

  • unclear risk; or

  • high risk (any non-random process, e.g. odd or even date of birth; hospital or clinic record number).

(2) Allocation concealment (checking for possible selection bias)

Was allocation adequately concealed?

For each included study, we planned to categorise the method used to conceal the allocation sequence as: 

  • low risk (e.g. telephone or central randomisation; consecutively numbered, sealed, opaque envelopes); 

  • unclear risk; or

  • high risk (open random allocation; unsealed or non-opaque envelopes, alternation; date of birth).

(3) Blinding (checking for possible performance bias)

Was knowledge of the allocated intervention adequately prevented during the study? At study entry? At the time of outcome assessment?

For each included study, we planned to categorise the methods used to blind study participants and personnel from knowledge of which intervention a participant received. We planned to assess blinding separately for different outcomes or classes of outcomes.

We planned to categorise the methods as: 

  • low risk, high risk, or unclear risk for participants; 

  • low risk, high risk, or unclear risk for outcome assessors;

  • low risk, high risk, or unclear risk for personnel.

(4) Incomplete outcome data (checking for possible attrition bias through withdrawals, drop-outs, protocol deviations)

Were incomplete outcome data adequately addressed?

For each included study, we planned to describe the completeness of data including attrition and exclusions from the analysis. We also planned to note the reason for attrition and exclusions if possible. 

We planned to categorise the methods as: 

  • low risk (< 20% missing data); 

  • unclear risk; or

  • high risk (≥ 20% missing data).

(5) Selective reporting bias

Were reports of the study free of suggestion of selective outcome reporting?

We planned to attempt to contact study authors, asking them to provide missing outcome data, when we suspected reporting bias. When this was not possible, and the missing data was thought to introduce serious bias, we planned to explore the impact of including such trials in the overall assessment of results by a sensitivity analysis.

For each included study, we planned to describe how we investigated the possibility of selective outcome reporting bias.

We planned to assess the methods as: 

  • low risk (where it is clear that all of the study's pre-specified outcomes and all expected outcomes of interest to the review have been reported); 

  • unclear risk; or

  • high risk (where not all the study's pre-specified outcomes have been reported).

(6) Other sources of bias

Was the study apparently free of other problems that could put it at a high risk of bias?

For each included study, we planned to describe any important concerns we had about other possible sources of bias (e.g. whether there was a potential source of bias related to the specific study design or whether the trial was stopped early due to some data-dependent process). We also planned to assess whether each study was free of other problems that could put it at risk of bias as:

  • low risk;

  • unclear risk; or

  • high risk.

Measures of treatment effect

We planned to use the standard methods of the Cochrane Neonatal Review Group. We planned to analyse categorical data using risk ratio (RR), risk difference (RD), and the number needed to treat for an additional beneficial outcome (NNTB). We also planned to analyse continuous data using the weighted mean difference (WMD) and report the 95% confidence interval (CI) for all estimates.

Assessment of heterogeneity

We planned to use the I2 statistic to measure heterogeneity among the trials in each analysis. We planned to explore it by pre-specified subgroup analysis, when we identified substantial heterogeneity (I2 statistic > 50%). In addition, we also planned to perform all statistical analyses using RevMan 5.1 (RevMan 2011) and Stata version 9.2 for Windows Stata.

Data synthesis

We planned to carry out statistical analysis using RevMan 5.1 (RevMan 2011). We also planned to use fixed-effect inverse variance meta-analysis for combining data where trials were examining the same intervention, and the trials' populations and methods are judged sufficiently similar. We planned to use fixed-effect meta-analysis where we could not explain heterogeneity between trials' treatment effects.

Results

Description of studies

Results of the search

From an initial search of 1447 citations, three studies were extracted for further examination. All three were excluded from the analysis (see "Characteristics of excluded studies" table below.

Included studies

None identified.

Excluded studies

Three studies identified but excluded.

Thompson 1986: One quasi-RCT (Thompson 1986) that met the participants and intervention criteria was excluded for reasons that we could not obtain adequate details of design and outcomes. We made efforts to contact the investigators with no success due to its old published year. No other RCTs and ongoing trials were identified.

This study was conducted in a tertiary children's hospital in the US. It was published in abstract form. Six infants with BPD were allocated to the hydralazine or placebo group in a blinded cross-over manner. It was unclear from the abstract if the study was randomised. Demographic and baseline parameters were as followings; mean body weight 910 ± 200 grams, gestational age 27 ± 2 weeks, postnatal age 57 ± 9 days, and FiO2 0.57 ± 0.12. Patients received either hydralazine 3.2 mg/kg/day or placebo orally for one week, no drug for one week, and the alternate drug for the third week. Echocardiogram, Doppler flow measurements, and pulmonary function studies were done at the beginning and end of the treatment period. Their primary and secondary outcomes were unclear from the abstract, and also they did not report mortality at 36 weeks' PMA, at hospital discharge, and at 18 and 36 months of age in either study group. RPET/RVET and AcT/RVET were calculated by echocardiogram and reported for the evaluation of PAH (Kosturakis 1984), but it was not presented with the data that we defined as PAH. Tidal volume, airway pressure, end-tidal PO2 and PCO2, dynamic compliance, total resistance, alveolar ventilation, and venous admixture were measured by pulmonary function test, but data were not presented in the abstract. They reported that AcT/RVET increased significantly with hydralazine, but other parameters were not different significantly between two groups; AcT/RVET increased from 0.18 to 0.23 (P = 0.01; only mean was presented), RPEP/RVET decreased from 0.32 to 0.27 (P = 0.07; only mean was presented). Four patients were extubated during the trial. They concluded that hydralazine did not cause hypoxaemia and did increase mix venous saturation, but again they did not report the exact data for them. There were no data reported on neurodevelopment assessment, length of hospitalisation, length of ventilation, length of oxygen supplementation, or adverse events.

Goodman 1988: Goodman studied the effects of oxygen and hydralazine on 15 infants with BPD and PAH in a well-organised non-controlled prospective manner. The haemodynamic responses to oxygen and hydralazine (0.3 mg/kg, via intravenous injection) were evaluated by cardiac catheterization. Five patients developed normal pulmonary artery pressure while receiving supplemental oxygen and were not studied further. Of the remaining 10 patients, the six patients with large, haemodynamically significant collateral vessels all had haemodynamically or clinically deleterious reactions to hydralazine. Two of the four patients without collateral pulmonary circulation responded to hydralazine with further reductions in mean pulmonary artery pressure to normal levels.

Martin 1991: Martin studied the effects of hydralazine for the cardiac performance in 23 infants with systemic hypertension undergoing extracorporeal membrane oxygenation (ECMO). Patients were divided into hypertensive group or normotensive group (40 mmHg < mean arterial pressure < 60 mmHg), and several parameters were measured on echocardiogram. The study was excluded because it was conducted in infants with respiratory failure on ECMO and did not address whether these infants were LBW or had CLD.

Ongoing studies

A search of PubMed, ClinicalTrials.gov, and the Clinical Trials Search Portal of the World Health Organization revealed no protocols or ongoing trials on our targeted patient group.

Risk of bias in included studies

We found no studies meeting the inclusion criteria for this review.

Effects of interventions

We found no studies meeting the inclusion criteria for this review.

Discussion

We found no studies that met our inclusion criteria for this review. Consequently, the review found no RCTs that addressed the use of hydralazine for LBW infant with persistent hypoxemic failure. Therefore, we could not establish if hydralazine treatment reduces mortality or neurodevelopmental impairment in our targeted patients.

Hydralazine has been used for the treatment of a variety of conditions. Although many newer drugs have been developed for the treatment of hypertension, hydralazine is still widely used in emergency and critical care fields due to its lower cost and extensive clinical benefits (Fivush 1997). Despite the benefits of hydralazine in hypertensive crisis and some other conditions (pre-eclampsia/eclampsia or chronic heart disease), it has not been widely used in PAH related to CLD. Following a comprehensive literature search, few clinical studies were identified that examined the effect of hydralazine in the management of PAH with CLD (Thompson 1986; Goodman 1988).

In infants with PAH related to CLD, Thompson (Thompson 1986) demonstrated that hydralazine administration may have potential benefit in PAH without causing hypoxia and haemodynamic instability. They alleged that pulmonary hypertension was improved with hydralazine by using data of RPEP/RVET and AcT/RVET (Kosturakis 1984). In this study, pulmonary function and mixed venous saturation were measured at the same time in order to demonstrate that no haemodynamic instability and complications occurred with hydralazine administration. But there are several potential limitations to this trial. First, only the abstract was obtained, so several points were unclear especially on the methodology. Although the patients' characteristics matched our protocol criteria in terms of prematurity (body weight (mean; 910 ± 200 g), gestational age (mean; 27 ± 2 weeks), and oxygen requirement (FiO2; 0.57 ± 0.12)), they did not report their exact definition of BPD and PAH, as well as other patient characteristics such as vital signs, medication used, and patients' oxygenation status. Second, even though the design of the trial was blind and cross-over manner, the number of the included patients was very small, and it was unclear if it was randomised or not. So strong selection bias need to be considered in this trial. Finally, they did not report any concrete data showing if there was any difference in pulmonary function test before and after hydralazine administration.

No other prospective controlled trials have been published on the use of hydralazine to treat PAH related to CLD. Moreover, only a few non-controlled trials are available, one of which showed potential adverse effects of hydralazine especially for the patients with pulmonary collateral shunt (Goodman 1988).

Although the studies suggested an potential improvement in PAH following infusion of hydralazine, there is no evidence from RCTs to support its use. Given the absence of evidence, hydralazine cannot be recommended as a treatment for PAH related to CLD. But since hydralazine is inexpensive and potentially beneficial for the patients group, RCTs are recommended. But such trials have to be carried out in settings where other treatments are not available, such as sildenafil, iNO, or extracorporeal life support, which are currently accepted in wealthy developed countries for the treatment of PAH, with careful monitoring of haemodynamics and complications.

Authors' conclusions

Implications for practice

There is insufficient evidence to determine the safety and efficacy of hydralazine in LBW infants with persistent hypoxemic respiratory failure.

Implications for research

A controlled clinical trial suggested a potential benefit of hydralazine in infants with persistent hypoxemic respiratory failure related to PAH and evolving CLD. Since hydralazine is inexpensive and potentially beneficial, RCTs are recommended. Such trials could be carried out in settings where other treatments such as sildenafil, iNO, or extracorporeal life support are not available, with careful monitoring of haemodynamics and complications.

Acknowledgements

None.

Data and analyses

Download statistical data

This review has no analyses.

Contributions of authors

Kawaguchi, Atsushi: conceptualised the review and protocol, literature search, reviewing the identified papers data extraction, drafting the manuscript.

Mori, Rintaro: contributed to design of the review and commented on the manuscript.

Isayama, Tetsuya: contributed to literature search, review of the identified papers, and data extraction.

Tamura, Masonori: commented on the manuscript.

Minami, Hirotaka: commented on the manuscript.

Yang, Ying: commented on the manuscript.

Declarations of interest

None.

Sources of support

Internal sources

  • No sources of support supplied

External sources

  • Eunice Kennedy Shriver National Institute of Child Health and Human Development National Institutes of Health, Department of Health and Human Services, USA.

    Editorial support of the Cochrane Neonatal Review Group has been funded with Federal funds from the Eunice Kennedy Shriver National Institute of Child Health and Human Development National Institutes of Health, Department of Health and Human Services, USA, under Contract No. HHSN275201100016C.

Differences between protocol and review

None.

Characteristics of studies

Characteristics of excluded studies [ordered by year of study]

StudyReason for exclusion
Thompson 1986Quasi-RCT that met the participants and intervention criteria; however, the study was excluded because we could not obtain adequate details regarding study design or relevant clinical outcomes
Goodman 1988This was a well-organised prospective study that evaluated the effects of oxygen and hydralazine to 15 BPD infants with pulmonary hypertension; however, intervention was not assigned in a random or quasi-random fashion
Martin 1991Studies the effects of hydralazine for systemic blood pressure in extracorporeal membrane oxygenation patients; no relevant outcomes were reported

Ancillary