Lavado pulmonar para el síndrome de aspiración de meconio en recién nacidos
Resumen
Antecedentes
El síndrome de aspiración de meconio (SAM) puede ocurrir cuando un recién nacido inhala una mezcla de meconio y líquido amniótico en sus pulmones cerca del momento del parto. Con excepción de las medidas de apoyo, existe poco tratamiento efectivo. El lavado pulmonar puede ser un tratamiento potencialmente efectivo para el SAM en virtud de eliminar el meconio de las vías respiratorias y alterar el curso natural de la enfermedad.
Objetivos
Evaluar los efectos del lavado pulmonar en la morbilidad y mortalidad en los recién nacidos con SAM.
Métodos de búsqueda
Se hicieron búsquedas en el Registro Cochrane Central de Ensayos Controlados (Cochrane Central Register of Controlled Trials) (CENTRAL, The Cochrane Library), MEDLINE, y en EMBASE hasta diciembre 2012; en las revisiones anteriores que incluyeron referencias cruzadas, resúmenes y actas de congresos; e informantes expertos. Se estableció contacto con los autores directamente para obtener datos adicionales. Se utilizaron los siguientes encabezados temáticos y palabras de texto: meconium aspiration, pulmonary surfactants, fluorocarbons, bronchoalveolar lavage, lung lavage, pulmonary lavage.
Criterios de selección
Ensayos controlados aleatorios que evaluaron los efectos del lavado pulmonar en neonatos con SAM, incluidos los que se intubaron para el lavado. El lavado pulmonar se definió como cualquier intervención en la que el líquido se instila en el pulmón, seguido de un intento de eliminarlo mediante la aspiración o el drenaje postural.
Obtención y análisis de los datos
Los revisores extrajeron de los informes del ensayo clínico los datos de los resultados clínicos, que incluyeron la mortalidad, el requisito para la oxigenación por membrana extracorpórea (OMEC), el neumotórax, la duración de la ventilación mecánica y la oxigenoterapia, la duración de la hospitalización, los índices de la función pulmonar y los efectos adversos del lavado. El análisis de los datos se realizó de acuerdo con las normas del Grupo Cochrane de Neonatología (Cochrane Neonatal Review Group).
Resultados principales
Sólo cuatro ensayos controlados aleatorios pequeños cumplieron con los criterios de selección. Para uno de estos ensayos, los datos del grupo de control no están disponibles. Dos estudios compararon el lavado mediante el agente tensioactivo diluido con la atención habitual. El metanálisis de estos dos estudios no mostró un efecto considerable en la mortalidad (riesgo relativo típico 0,42; intervalo de confianza [IC] del 95%: 0,12 a 1,46; diferencia de riesgos típica ‐0,10; IC del 95%: ‐0,24 a 0,04) o la administración de OMEC (riesgo relativo típico 0,27; IC del 95%: 0,04 a 1,86; diferencia de riesgos típica ‐0,15; IC del 95%: ‐0,35 a 0,04). Para el resultado compuesto de mortalidad o el uso de la OMEC, un efecto considerable favoreció el grupo de lavado (riesgo relativo típico 0,33; IC del 95%: 0,11 a 0,96; diferencia de riesgos típica ‐0,19; IC del 95%: ‐0,34 a ‐0,03; número necesario a tratar [NNTB] 5). No se informaron otros beneficios. El otro estudio que se publicó comparó el lavado con agente tensioactivo seguido de un surfactante administrado en bolo con el tratamiento con bolo de surfactante solo en el SAM complicado por hipertensión pulmonar. No se observaron mejorías significativas en la mortalidad, el neumotórax, la duración de la ventilación mecánica o la duración de la hospitalización.
Conclusiones de los autores
En los neonatos con síndrome de aspiración de meconio, el lavado pulmonar con agente tensioactivo diluido puede ser beneficioso, pero los ensayos clínicos controlados adicionales del tratamiento con lavado deben realizarse para confirmar el efecto del tratamiento, para refinar el método de tratamiento con lavado y para comparar el tratamiento con lavado con otros abordajes, incluido el tratamiento de surfactante administrado en bolo. Los resultados a largo plazo deben evaluarse en ensayos clínicos adicionales.
PICO
Resumen en términos sencillos
Lavado pulmonar para el síndrome de aspiración de meconio en recién nacidos
El síndrome de aspiración de meconio (SAM) es una enfermedad pulmonar del recién nacido en la que el meconio, heces fetales, pasa antes del nacimiento y luego se respira en los pulmones. Aparte de las medidas de apoyo como la respiración artificial y, ocasionalmente, la derivación cardiopulmonar, existe poco tratamiento efectivo. Esta revisión evaluó si la limpieza pulmonar mediante un agente tensioactivo llamado químico natural, u otro líquido similar, es útil en el SAM. Este procedimiento de limpieza se conoce como lavado pulmonar. El lavado pulmonar con agente tensioactivo diluido puede ayudar a mejorar el curso clínico de los neonatos con SAM, en particular, la probabilidad de la supervivencia sin la necesidad de derivación cardiopulmonar. Se necesitarán más ensayos para evaluar adecuadamente el tratamiento con lavado en el SAM.
Authors' conclusions
Background
Description of the condition
Meconium aspiration syndrome (MAS) occurs when a newborn infant inhales a mixture of meconium and amniotic fluid into the lungs around the time of delivery. Intrapartum passage of meconium, the viscid secretion of the fetal intestine, occurs in up to 15% of deliveries at term (Wiswell 1993). Aspiration of meconium or meconium‐stained amniotic fluid into the airways may occur prenatally (during hypoxic fetal gasping), or immediately after delivery as the first breaths are taken. Once inhaled, meconium migrates down the tracheobronchial tree, causing a variable degree of airway obstruction as it disperses into the distal airspaces (Tran 1980). Thereafter a toxic pneumonitis ensues, with hemorrhagic edema and exudation of plasma proteins into the alveolar space (Tyler 1978; Dargaville 2001). The function of pulmonary surfactant may be secondarily impaired, both by meconium (Moses 1991), and by plasma protein (Fuchimukai 1987). In many infants with MAS, particularly those with coexisting asphyxia, there is an added component of pulmonary hypertension, which may cause profound hypoxaemia. The reported incidence of MAS varies widely, but is of the order of 1 to 2 per 1000 live births (Wiswell 1993).
Approximately one‐third of infants with MAS need mechanical ventilatory support (Wiswell 1993), and many are treated with high‐frequency ventilation or nitric oxide, or both. Infants ventilated for MAS are often treated with exogenous surfactant, which appears to reduce the use of extracorporeal membrane oxygenation (ECMO), but has no clear effect on mortality (Findlay 1996; Lotze 1998; El Shahed 2007). Data regarding the effect of bolus surfactant therapy on pulmonary complications of MAS are conflicting; one small trial demonstrated a benefit in terms of air leak and duration of ventilation (Findlay 1996), but further studies have revealed no evidence of an effect on pulmonary complications (Lotze 1998, Chinese Collaborative Study Group 2005; Maturana 2005). Meta‐analysis of the data from these trials supports reduction in the use of ECMO, but not a reduced incidence of pulmonary complications of MAS (El Shahed 2007).
Description of the intervention
In human infants, the history of lavage as a therapy for MAS extends back to the early 1970s, when saline lavage was used in the delivery room to improve clearance of meconium from the airways of meconium‐stained babies (Burke‐Strickland 1973). This technique was largely abandoned as a result of the increase in numbers of infants with transient tachypnoea after saline lavage, ascribed to lavage fluid retention in the lung (Carson 1976). Isolated reports of lung lavage in infants with established MAS subsequently appeared (Ibara 1995; Mosca 1996), in which lavage with a total volume of 20 to 40 mL/kg saline was performed, followed by bolus administration of natural surfactant. Improvement in oxygenation and carbon dioxide (CO2) clearance was noted in each case. Several research groups have reported their experience of lavage with dilute surfactant in ventilated infants with MAS, on the whole suggesting improvements in oxygenation and/or duration of ventilation in comparison with historical or concurrent controls (Su 1998; Lam 1999; Kowalska 2002; Schlösser 2002; Chang 2003; Salvia‐Roigés 2004; Dargaville 2007).
How the intervention might work
Recent experimental studies have suggested that lung lavage can remove meconium from the lungs in MAS, and as a result can improve lung function. In animal models of MAS, lung lavage using total fluid volumes of 10 to 60 mL/kg has resulted in considerable improvement in oxygenation and/or pulmonary mechanics, associated with removal of one‐third to one‐half of the meconium lodged in the airspaces (Paranka 1992; Ohama 1994; Cochrane 1998; Ohama 1999; Dargaville 2003). Saline, surfactant, and perfluorocarbon have been studied as potential lavage fluids. Comparative data suggest that exogenous surfactant, whether at full strength (Paranka 1992) or diluted in saline (Ohama 1994; Cochrane 1998; Ohama 1999), is a more effective lavage fluid than saline alone, in terms of both pulmonary function post lavage and removal of meconium from the lung. Lavage with perfluorocarbon appears to be superior to saline lavage (Marraro 1998) but less effective than dilute surfactant lavage (Dargaville 2003). The volume of each lavage aliquot is another determinant of lavage efficacy, with aliquot volumes of 15 mL/kg being more effective than multiple 2‐ to 3‐mL aliquots (Dargaville 2003), or aliquots of 8 mL/kg (Dargaville 2008).
Why it is important to do this review
At present, the therapeutic emphasis in MAS is on providing supportive care, with little or no effort directed towards removal of meconium from the lung as a means of halting disease progression. Recent data suggest that meconium can be safely removed from the airspaces in MAS by lung lavage. The objective of this review is to critically appraise the data from controlled trials of lavage therapy in human infants with MAS, and thereby evaluate the efficacy and safety of lung lavage as a treatment for this disease. The following systematic review evaluates randomised controlled trials that have studied the efficacy of lung lavage therapy in infants with MAS.
Objectives
To evaluate the effects of lung lavage on morbidity and mortality in newborn infants with MAS.
Subgroup analyses: to evaluate the effects of the type of lavage fluid, the volume of lavage fluid, and the timing of administration of lavage fluid on morbidity and mortality in newborn infants with MAS.
Methods
Criteria for considering studies for this review
Types of studies
All randomised or quasi‐randomised studies comparing therapeutic lung lavage with standard care in the management of infants with MAS.
Types of participants
Newborn infants with MAS (infants delivered through meconium‐stained amniotic fluid with early onset of respiratory distress, no other obvious cause for the distress, and either a characteristic chest X‐ray or meconium found beyond the vocal cords at or after delivery). This review includes infants already intubated at the time of lavage, and infants intubated for the purpose of lavage.
Types of interventions
Lung lavage is defined as any intervention wherein fluid is instilled into the lung, followed by an attempt to remove it by suctioning, postural drainage, or both. Fluids that have been used for this purpose include saline, full‐strength and dilute surfactant, and perfluorocarbon.
Standard care is defined as no lavage therapy, but it does include routine suction of the endotracheal tube to maintain its patency. For some studies, bolus surfactant therapy may be mandated as part of standard care.
Types of outcome measures
Primary outcomes
-
Death.
-
Use of ECMO.
-
Death or use of ECMO.
-
Pneumothorax.
-
All air leak (pneumothorax, pneumomediastinum, pneumopericardium, pneumoperitoneum, pulmonary interstitial emphysema).
-
Days of mechanical ventilation via an endotracheal tube.
-
Days of supplemental oxygen.
-
Length of stay in hospital.
-
Total cost of hospitalisation.
The composite outcome of death or use of ECMO has been included in recognition that mortality is influenced by the availability of ECMO.
Secondary outcomes
-
Indices of pulmonary function (including Oxygenation Index, Alveolar‐arterial oxygen difference, PF ratio) measured at 24, 48, and 72 hours.
[definitions: Oxygenation index (OI) = (Mean airway pressure × FiO2) / PaO2; Alveolar‐arterial oxygen difference (AaDO2) = FiO2 × 713 － PaCO2/0.8 － PaO2; PF Ratio = PaO2/FiO2]
Lung mechanics (compliance and resistance of the lung or the respiratory system):
-
Adverse effects of lavage (acute hypoxaemia, bradycardia, hypotension).
Search methods for identification of studies
We used the standard search methods of the Cochrane Neonatal Review Group.
Electronic searches
We used the standardized search strategy of the Neonatal Review Group as outlined in The Cochrane Library. The following sources were searched between 1966 and December 2012 for eligible studies in any language:
-
Cochrane Neonatal Review Group trials register.
-
CENTRAL (The Cochrane Library, Issue 11, 2012).
-
MEDLINE and EMBASE electronic searches.
We constructed search strategies using the following MeSH terms or keywords: meconium, meconium aspiration syndrome, pulmonary surfactants, lung surfactant, fluorocarbons, bronchoalveolar lavage, lung lavage, and pulmonary lavage.
Searching other resources
We screened for trials in conference proceedings of annual meetings of the American Thoracic Society, the Society for Pediatric Research, the European Respiratory Society, and the European Society for Pediatric Research (December 2012); and in the reference lists from the retrieved articles and from review articles. We had personal communications with primary authors of the identified studies to identify unpublished data.
We searched for any ongoing or recently completed and unpublished trials using clinicaltrials.gov, controlled‐trials.com, and who.int/ictrp.
Data collection and analysis
We used the methods of the Cochrane Neonatal Review Group for data collection and analysis.
Selection of studies
We included all randomised and quasi‐randomised controlled trials that fulfilled the selection criteria described in the previous section. Two review authors independently reviewed the results of the updated search and selected studies for inclusion. We resolved any disagreement by discussion.
Data extraction and management
We used a standard form for data extraction, which included a collection of descriptive data on the study design, the study population (baseline characteristics and inclusion and exclusion criteria), and the method of intervention (type of lavage fluid, total lavage volume, aliquot volume, and lavage fluid concentration), and quantitative data regarding the outcome measures. Pneumothorax was counted only if it occurred after randomisation.
Two review authors independently extracted the data from included studies, and results were compared. The investigators of included studies were asked to provide unpublished outcome data where necessary.
Assessment of risk of bias in included studies
The quality of eligible studies was assessed using The Cochrane Collaboration’s tool for assessing the risk of bias for randomised controlled trials (RCTs) (Higgins 2011). Two review authors performed the assessment, and they resolved any difference of opinion by involving coauthors in the discussion.
The methodological quality of the studies was assessed using the following criteria:
-
Sequence generation (checking for possible selection bias): For each included study, we categorized the method used to generate the allocation sequence as follows:
-
Low risk (any truly random process, e.g. random number table; computer random number generator).
-
High risk (any nonrandom process, e.g. odd or even date of birth; hospital or clinic record number).
-
Unclear risk.
-
-
Allocation concealment (checking for possible selection bias): For each included study, we categorized the method used to conceal the allocation sequence as follows:
-
Low risk (e.g. telephone or central randomisation; consecutively numbered sealed opaque envelopes).
-
High risk (open random allocation; unsealed or non‐opaque envelopes, alternation; date of birth).
-
Unclear risk.
-
-
Blinding (checking for possible performance bias): For each included study, we categorized the methods used to blind study participants and personnel from knowledge of which intervention a participant received. Blinding was assessed separately for different outcomes or classes of outcomes. We categorized the methods as follows:
-
Low risk, high risk, or unclear risk for participants.
-
Low risk, high risk, or unclear risk for personnel.
-
Low risk, high risk, or unclear risk for outcome assessors.
-
-
Incomplete outcome data (checking for possible attrition bias through withdrawals, dropouts, and protocol deviations): For each included study and for each outcome, we described the completeness of data, including attrition and exclusions from the analysis. We noted whether attrition and exclusions were reported, the numbers included in the analysis at each stage (compared with the total randomly assigned participants), reasons for attrition or exclusion where reported, and whether missing data were balanced across groups or were related to outcomes. Where sufficient information was reported or supplied by the trial authors, we included missing data in the analyses again. We categorized the methods as follows:
-
Low risk (< 20% missing data).
-
High risk (≥ 20% missing data).
-
Unclear risk.
-
-
Selective reporting bias: For each included study, we described how we investigated the possibility of selective outcome reporting bias and what we found. We assessed the methods as follows:
-
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).
-
High risk (where not all of the study’s pre‐specified outcomes have been reported; one or more reported primary outcomes were not pre‐specified; outcomes of interest are reported incompletely and so cannot be used; or study fails to include results of a key outcome that would have been expected to have been reported).
-
Unclear risk.
-
-
Other sources of bias: For each included study, we described any important concerns that we had about other possible sources of bias (e.g. whether a potential source of bias was related to the specific study design, or whether the trial was stopped early as the result of some data‐dependent process). We assessed whether each study was free of other problems that could put it at risk of bias as follows:
-
Low risk; high risk; unclear risk.
-
-
Overall risk of bias [described in Table 8.5c in the Handbook].
We made explicit judgements regarding whether studies were at high risk of bias, according to the criteria given in The Cochrane Handbook (Higgins 2011). With reference to (1) to (6) above, we assessed the likely magnitude and direction of the bias, and whether we considered it likely to influence the findings. If needed, we planned to explore the impact of the level of bias by undertaking sensitivity analyses (see Sensitivity analysis, later).
Measures of treatment effect
We performed statistical analyses using Review Manager software (RevMan 2011). Dicotomous data were analysed using relative risk (RR), risk difference (RD), and the number needed to benefit (NNTB) or the number needed to harm (NNTH). The 95% confidence intervals (CIs) were reported on all estimates.
Some continuous outcomes are only descriptively presented in a table without statistical pooling because of the skewed nature of the data; the weighted mean difference (WMD) was used for pooling otherwise.
Dealing with missing data
For included studies, levels of attrition were noted. The impact of including studies with high levels of missing data in the overall assessment of treatment effect was explored through sensitivity analysis.
All outcome analyses were performed on an intention‐to‐treat basis(i.e. we included all participants randomly assigned to each group in the analyses). The denominator for each outcome in each trial was the number randomly assigned minus any participants whose outcomes were known to be missing.
Assessment of heterogeneity
We examined heterogeneity between trials by inspecting the forest plots and quantifying the impact of heterogeneity using the I2 statistic. If noted, we planned to explore the possible causes of statistical heterogeneity using pre‐specified subgroup analyses (e.g. differences in study quality, participants, intervention regimens, or outcome assessments).
Assessment of reporting biases
We planned to assess possible publication bias and other biases using symmetry/asymmetry of funnel plots, but this was not applicable because an insufficient number of studies was included in the meta‐analysis for such an exploration.
For included trials that were recently performed (and therefore were prospectively registered), we explored possible selective reporting of study outcomes by comparing primary and secondary outcomes given in the reports versus primary and secondary outcomes proposed at trial registration, using the Websites www.clinicaltrials.gov and www.controlled‐trials.com. If such discrepancies were found, we planned to contact the primary investigators to obtain missing outcome data on outcomes pre‐specified at trial registration.
Data synthesis
Where meta‐analysis was judged to be appropriate, the analysis was done using Review Manager software (RevMan 2011), as supplied by The Cochrane Collaboration. We used the Mantel‐Haenszel method to obtain estimates of typical relative risk and risk difference. A fixed‐effect model was primarily used for the meta‐analysis after the statistical heterogeneity was investigated. Data were analysed on an intention‐to‐treat basis.
Subgroup analysis and investigation of heterogeneity
Three subgroup analyses were planned a priori:
-
Type of lavage fluid (saline, surfactant, perfluorocarbon, other).
-
Lavage aliquot volume (< 5 mL/kg, ≥ 5 mL/kg).
-
Timing of lavage: early (< 6 hours of life) or late (≥ 6 hours of life).
Sensitivity analysis
We planned sensitivity analyses for use in situations where this might affect the interpretation of significant results (e.g. where risk of bias is associated with the quality of some of the included trials or missing outcome data). None were thought necessary in this review.
Results
Description of studies
Results of the search
Four randomised controlled trials were identified, one of which (Ogawa 1997) was excluded as data on the nonlavaged control group were not reported and are not now obtainable. In that study, which has been published only in conference proceedings format, six infants underwent lavage with five aliquots each of 2 mL/kg of Surfacten‐TA (6 mg/mL), and a further four infants received an identical lavage using saline. Oxygenation and CO2 clearance were better in the group lavaged with dilute surfactant than with saline, but no formal comparisons with the control group are reported.
Included studies
Three studies are included in this review (Wiswell 2002; Gadzinowski 2008; Dargaville 2011).
Wiswell 2002 performed a phase I/II randomised controlled trial of surfactant lavage in conventionally ventilated infants with MAS who were at least 35 weeks' gestation and less than 72 hours of age, and had an oxygenation index (OI) between 8 and 25 inclusive on two separate blood gas analyses within a three‐hour period. Surfactant lavage was performed at a mean age of 14 hours using 6 aliquots of 8 mL/kg of KL4 (Surfaxin, Discovery Laboratories Inc, Doylestown, PA). The concentration of surfactant phospholipid was 2.5 mg/mL for the first four aliquots, and 10 mg/mL for the last two. Each lavage aliquot was instilled via the endotracheal tube while positive end‐expiratory pressure continued, followed by closed endotracheal suctioning for 10 seconds, during which positive‐pressure ventilation was re‐instituted. The infant's physiological state was allowed to recover after each lavage aliquot before the next was administered. After the final aliquot, positive end‐expiratory pressure was maintained at 6 to 8 cm H2O for at least two hours. Control infants received conventional mechanical ventilation and standard supportive measures at the discretion of the site study investigator. In both groups, a treatment failure criterion (OI > 25 or OI 50% above baseline) had to be reached before rescue therapies such as high‐frequency oscillatory ventilation (HFOV), bolus surfactant therapy, inhaled nitric oxide (iNO), and ECMO could be used.
Dargaville 2011 performed a multicenter randomised controlled trial of diluted surfactant lavage in infants who had a diagnosis of MAS. The infants were at least 36 weeks' gestation and 2 kg birth weight, less than 24 hours of age, and mechanically ventilated with a mean airway pressure of at least 12 cm H2O and an alveolar‐arterial O2 difference (AaDO2) of at least 450 mmHg on two sequential blood gases. Surfactant lavage was performed at a mean age of 13 hours using two aliquots of 15 mL/kg of bovine surfactant (Survanta, Abbott Laboratories, Columbus, OH) diluted with saline to a phospholipid concentration of 5 mg/mL. Lavage fluid was instilled over 20 seconds through a dispensing catheter with the ventilator circuit disconnected. Three positive‐pressure inflations were then administered, followed by disconnection of the ventilator circuit and suctioning of the instilled fluid. Control infants received mechanical ventilation and standard supportive measures. In both groups, ventilator management and the use of HFOV, iNO, and bolus surfactant therapy were at the discretion of the site study investigator, as was the decision to refer to ECMO.
Gadzinowski 2008 performed a randomised controlled trial of surfactant lavage followed by bolus surfactant treatment compared with bolus surfactant treatment alone for MAS with pulmonary hypertension. The infants were at least 35 weeks' gestation and less than 24 hours of age, and the diagnosis of pulmonary hypertension was based on standardised echocardiographic parameters. Surfactant lavage was performed with a total lavage volume of 15 mL/kg and an aliquot volume of 3.75 mL/kg at the mean age of 9.7 hours, using diluted bovine surfactant (Survanta) at a phospholipid concentration of 5 mg/mL. Lavage and suctioning were conducted via a closed system in four body positions: on the right and left sides, and in the Trendelenburg and anti‐Trendelenburg positions. After the lavage treatment, one dose of bolus surfactant (Survanta, 100 mg/kg) was given. The control group received one dose of bolus surfactant (Survanta, 100 mg/kg) and conventional treatment. After an echocardiographic assessment was conducted, iNO was administered to both groups.
Excluded studies
Twelve studies were excluded from the analysis (Burke‐Strickland 1973; Carson 1976; Rosegger 1987; Ogawa 1997; Su 1998; Lam 1999; Schlösser 2002; Kowalska 2002; Chang 2003; Salvia‐Roigés 2004; Dargaville 2007; Armenta 2011). The rationale for exclusion is given in the table Characteristics of excluded studies.
Ongoing studies
Ongoing or unpublished trials are noted in the table Characteristics of ongoing studies (McNamara 2006; Segal 2012; Sur‐Lu‐Lav 2011).
Risk of bias in included studies
In the study by Wiswell et al (Wiswell 2002), randomisation was performed by drawing a randomisation slip from a closed envelope, with an allocation ratio of 2:1 (lavage:standard care). Treatment was not blinded to the clinical team, although the one‐year follow‐up was performed by an investigator who was blinded to the allocation. No exclusions were noted after randomisation, although three of the 15 infants randomly assigned to surfactant lavage did not receive the complete lavage series, two infants received only four of the scheduled six lavage aliquots, and another received only two aliquots. For the purposes of analysis, all infants were included in their respective allocation groups. Other bias may have existed in that the number of infants receiving rescue therapy exceeded the number reaching treatment failure criteria, even though rescue therapies were not permitted unless infants met treatment failure. The study was conducted without a formal sample size calculation and based on an estimate for assessing safety and potential efficacy in a rather exploratory fashion.
The study by Dargaville et al (Dargaville 2011) described an adequate process of randomisation and allocation concealment. Among 66 enrolled infants, one infant randomly assigned to the surfactant lavage group was too unstable to receive lavage and was deemed to have been ineligible for enrolment. The intervention was not blinded.
In the study by Gadzinowski et al (Gadzinowski 2008), no information is provided about random sequence generation, allocation concealment, and blinding of the intervention.
Effects of interventions
LUNG LAVAGE VERSUS STANDARD CARE (Comparison 1)
Two studies compared lung lavage with standard care (Dargaville 2011; Wiswell 2002).
Death (Outcome 1.1)
Both studies reported on mortality, and one RCT reported no events. No treatment effect on death was noted (typical RR 0.42, 95% CI 0.12 to 1.46; typical RD ‐0.10, 95% CI ‐0.24 to 0.04) (Analysis 1.1).
Use of ECMO (Outcome 1.2)
Both RCTs reported on the number of infants who needed ECMO. In one study (Dargaville 2011), only 25 of the 66 enrolled infants were treated at centres at which ECMO was available. No difference in the relative risk of ECMO was noted, although a trend toward an interventional benefit was observed (typical RR 0.27, 95% CI 0.04 to 1.86; typical RD ‐0.15, 95% CI ‐0.35 to 0.04) (Analysis 1.2).
Death or use of ECMO (Outcome 1.3)
For both studies, the numbers of infants who received ECMO or died could be calculated. Surfactant lavage significantly decreased the combined outcome of death or requirement for ECMO (typical RR 0.33, 95% CI 0.11 to 0.96: typical RD ‐0.19, 95% CI ‐0.34 to ‐0.03; NNTB 5) (Analysis 1.3). Each study showed a result favouring the intervention.
Pneumothorax (Outcome 1.4)
Both studies reported on pneumothorax, but not on other air leaks. No significant difference was observed between treatment groups (typical RR 0.38, 95% CI 0.08 to 1.90, typical RD ‐0.07, 95% CI ‐0.19 to 0.05) (Analysis 1.4).
Days of mechanical ventilation
Duration of ventilation showed wide variation in both studies. Median duration for mechanical ventilation was shorter in the intervention group for both studies (Dargaville 2011); median 5.0 versus 6.3 days (Wiswell 2002); median 4.6 versus 7.6 days) (Table 1).
| Study | Number of infants (i/c) | Days of mechanical ventilation | Days of supplemental oxygen | Length of hospital stay (days) | |||
| Intervention | Control | Intervention | Control | Intervention | Control | ||
| 31/35 | 5.0 (3.3‐8.7) | 6.3 (3.9‐8.1) | 14 (6.7‐21) | 14 (11‐18) | 17 (11‐25) | 19 (15‐25) | |
| 7/15 | 4.6 (1.1‐22.3) | 7.6 (1.1‐28) | 13.5 ± 9.3 | 12.1 ± 10.7 | 12.7 ± 8.7 | 13.1 ± 10.3 | |
i = intervention group; c = control group.
All variables expressed by mean ± standard deviation or median (interquartile range). Data from the survivors were used.
Days of supplemental oxygen
Dargaville 2011 reported days of oxygen therapy only for survivors.The values were a little different between groups: (Dargaville 2011): median 14 versus 14 days (Wiswell 2002): mean 13.5 versus 12.1 days (Table 1).
Length of hospital stay
One study (Wiswell 2002) reported length of stay in the Neonatal Intensive Care Unit, which differed little between groups (mean 12.7 vs 13.1 days). In the other study, the length of hospital stay was similar in the two groups (Dargaville 2011): median 17 versus 19 days (Table 1).
Total cost of hospitalisation
None of the studies reported hospitalisation cost.
Indices of pulmonary function (Outcomes 1.5 and 1.6)
Both RCTs reported OI measured at 24, 48, and 72 hours. A significant difference between groups was observed at 48 hours after lavage treatment (WMD ‐6.20, 95% CI ‐12.11 to ‐0.29) (Analysis 1.5). AaDO2 and pulmonary function (PF) ratio were measured in one study (Dargaville 2011), which did not show any significant differences between groups, although better results appeared to be obtained in the treatment group over time after use of lavage therapy (Analysis 1.6). Lung mechanics were not reported in either study.
Adverse effects
In one study (Wiswell 2002), the instillation and recovery of the six lavage aliquots took 50 to 60 minutes. In two infants the hypoxaemia that occurred during lavage was sufficiently pronounced to halt the lavage procedure. Overall 5 of 15 infants required hand ventilation to recover oxygen saturation after lavage. In one other infant, the lavage procedure was stopped because of hypotension, although this infant had coincident gram‐negative sepsis. The occurrence and severity of episodes of hypoxaemia or hypotension are not reported in the control group, and thus it is not possible to make direct comparisons between groups. In the other study (Dargaville 2011), two infants experienced transient bradycardia at less than 100 beats per minute during lavage, with recovery by five minutes after lavage. Five infants had an oxygen saturation below 80% for longer than 10 minutes, with recovery to above 90% within 40 minutes in all cases. Six infants needed treatment for hypotension during or immediately after lavage. Overall, cardiopulmonary indices were affected transiently, and the lavaged infants and the control infants showed similar blood gas indices at four hours post lavage. One infant died of intractable pulmonary hypertension three hours after lavage.
LUNG LAVAGE FOLLOWED BY SURFACTANT BOLUS VERSUS SURFACTANT BOLUS THERAPY FOR MAS WITH PULMONARY HYPERTENSION (Comparison 2)
One study compared lung lavage followed by surfactant bolus versus surfactant bolus therapy for MAS with pulmonary hypertension (Gadzinowski 2008).
Death (Outcome 2.1)
No difference in the relative risk of mortality was noted; two deaths were reported in the control group versus none in the lavage group (RR 0.17, 95% CI 0.01 to 3.06) (Analysis 2.1).
Pneumothorax (Outcome 2.2)
No difference in the relative risk of pneumothorax was noted; two episodes of pneumothorax were reported in the control group but none in the lavage group (RR 0.17, 95% CI 0.01 to 3.06) (Analysis 2.2).
Days of mechanical ventilation
The difference between mean values in days of mechanical ventilation was less than one day (mean ± standard deviation [SD] 6.6 ± 2.6 vs 7.3 ± 1.7 days) (Table 2).
| Study | Number of infants (i/c) | Days of mechanical ventilation | Length of hospital stay (days) | ||
| Intervention | Control | Intervention | Control | ||
| 7/6 | 6.6 ± 2.6 | 7.3 ± 1.7 | 16.4 ± 5.4 | 19.8 ± 2.9 | |
i = intervention group; c = control group.
All variables expressed by mean ± standard deviation.
Length of the hospital stay
The length of hospital stay appeared to be shorter in the intervention group (mean ± SD: 16.4 ± 5.4 vs 19.8 ± 2.9 days) (Table 2).
Indices of pulmonary function
As a result of the small size of the study (7 vs 6 for treatment vs control) and the skewed nature of the data (large difference between reported mean and median values), we did not assess the significance based on the test of means but just descriptively presented the results (Table 3). The median value of OI measured at 24 hours in the surfactant lavage group was lower than that in the control group (2.8 vs 9.0). The OI measured at 48 hours was, however, similar between groups (median 1.7 vs 1.8). AaDO2 measured at 24 and 48 hours in the lavage group appeared to be lower than in the control group. Compliance and resistance are not reported.
|
| Intervention (n = 7) | Control (n = 6) |
| Oxygenation index (0 hours) | 29.8 ± 12.5 (median 25) | 32.4 ± 25 (median 24.3) |
| Oxygenation index (24 hours) | 2.7 ± 2.2 (median 2.8) | 10.4 ± 8.1 (median 9.0) |
| Oxygenation index (48 hours) | 5.0 ± 9.1 (median 1.7) | 5.7 ± 9.1 (median 1.8) |
| AaDO2 (0 hours) | 575.6 ± 91.0 | 589.5 ± 95.8 |
| AaDO2 (24 hours) | 261.0 ± 160.9 | 352.3 ± 177.5 |
| AaDO2 (48 hours) | 178.3 ± 144.0 | 226.0 ± 239.8 |
i = intervention group; c = control group; AaDO2 = alveolar‐arterial oxygen difference (mmHg),
All variables expressed by mean ± standard deviation or median.
Adverse effects
Except for pneumothorax and death, adverse effects were not reported.
SUBGROUP ANALYSES
None of the planned subgroup analyses were possible.
Type of lavage fluid
All included studies used diluted surfactant for lavage.
Lavage aliquot volume
The aliquot volume was at least 5 mL/kg in all studies comparing surfactant lavage with standard care, and less than 5 mL/kg in the study comparing surfactant lavage followed by bolus surfactant with surfactant bolus therapy.
Timing of lavage
The mean age when lavage was performed was greater than six hours in all included studies.
Discussion
Therapeutic lung lavage is an emerging treatment for MAS, which, by virtue of removal of meconium from the lung, would appear to have a potential advantage over the supportive measures currently employed for this condition. This review has identified three small randomised controlled trials of lung lavage using surfactant (Wiswell 2002; Gadzinowski 2008; Dargaville 2011).
In the meta‐analysis of the trials comparing surfactant lavage and standard care (Wiswell 2002; Dargaville 2011), a significant difference was noted in the composite outcome of death or use of ECMO. Analysis of this outcome was necessary given that the availability of ECMO clearly affects mortality. Any other primary outcomes including mortality, pneumothorax, or use of ECMO did not demonstrate a significant benefit. Among the secondary outcomes examining pulmonary function, only OI at 48 hours was improved significantly in the surfactant lavage group. In one study in which a large total volume of lavage fluid was used, the lavage procedure was relatively protracted and in some cases was halted because of concern regarding hypoxaemia or hypotension (Wiswell 2002). In the other study, the lavage procedure was completed in all infants, but some experienced transient bradycardia and hypotension.
The two studies comparing surfactant lavage with standard care varied considerably in the severity of disease of enrolled infants at the time of recruitment (Wiswell 2002; Dargaville 2011). In the study of Wiswell et al, infants with MAS of lesser severity were targeted (mean OI 12 at enrolment), and no deaths and relatively rapid weaning from ventilation were noted, in particular in the lung lavage group. By contrast, the other study focused on infants with severe disease (mean OI 25 at enrolment) (Dargaville 2011). In this case, no difference was discernible in duration of mechanical ventilation, which was relatively prolonged in both groups, but fewer infants who underwent lavage died or required ECMO. This suggests that lung lavage has the greatest potential for benefit in infants with severe disease, although the possibility of an impact in milder cases on duration of ventilation or other pulmonary outcomes needs further exploration.
The study comparing surfactant lavage followed by bolus surfactant with surfactant bolus therapy (Gadzinowski 2008) did not show an effect on mortality, pneumothorax, days on mechanical ventilation, or length of hospital stay. The intervention seemed to improve oxygenation, with a lower OI at 24 hours.
Because of the small number of RCTs and the lack of sufficient numbers of infants randomly assigned, the evidence regarding lung lavage in MAS is thought to be insufficient to allow firm conclusions, although a beneficial effect is noted in some important outcomes. A recent systematic review focusing on surfactant lavage therapy reviewed existing RCTs together with non‐randomised controlled studies for supporting evidence; the results of meta‐analysis also suggested that surfactant lavage had significant effects on mortality and morbidity for MAS (Choi 2012).
Subgroup analyses according to type of lavage fluid, lavage aliquot volume, and timing of lavage were planned. However, none of the planned subgroup analyses were possible. Additional studies will be needed to fully assess the impact of these factors on the success of lung lavage in MAS.
Of the few RCTs identified, only one was judged to have a low risk of bias. The protocols of the other two studies were unavailable for full assessment.
Further randomised controlled trials of lung lavage are needed to properly evaluate the safety and efficacy of this treatment. A phase III RCT (Segal 2012) evaluating the effect of surfactant lavage compared with standard care had been registered, and is recorded to have been terminated without completion. One trial comparing lavage with diluted surfactant versus standard care is currently under way (McNamara 2006) and is aiming to recruit 60 infants. Another trial (Sur‐Lu‐Lav 2011) undertaken to investigate the effect of surfactant lavage compared with standard care is registered.
Comparison 1 Lung lavage versus standard care, Outcome 1 Death.
Comparison 1 Lung lavage versus standard care, Outcome 2 Use of ECMO.
Comparison 1 Lung lavage versus standard care, Outcome 3 Death or use of ECMO.
Comparison 1 Lung lavage versus standard care, Outcome 4 Pneumothorax.
Comparison 1 Lung lavage versus standard care, Outcome 5 Oxygenation index.
Comparison 1 Lung lavage versus standard care, Outcome 6 Alveolar‐arterial oxygen difference.
Comparison 1 Lung lavage versus standard care, Outcome 7 PaO2/FiO2.
Comparison 2 Lung lavage followed by surfactant bolus versus surfactant bolus, Outcome 1 Death.
Comparison 2 Lung lavage followed by surfactant bolus versus surfactant bolus, Outcome 2 Pneumothorax.
| Study | Number of infants (i/c) | Days of mechanical ventilation | Days of supplemental oxygen | Length of hospital stay (days) | |||
| Intervention | Control | Intervention | Control | Intervention | Control | ||
| 31/35 | 5.0 (3.3‐8.7) | 6.3 (3.9‐8.1) | 14 (6.7‐21) | 14 (11‐18) | 17 (11‐25) | 19 (15‐25) | |
| 7/15 | 4.6 (1.1‐22.3) | 7.6 (1.1‐28) | 13.5 ± 9.3 | 12.1 ± 10.7 | 12.7 ± 8.7 | 13.1 ± 10.3 | |
| i = intervention group; c = control group. All variables expressed by mean ± standard deviation or median (interquartile range). Data from the survivors were used. | |||||||
| Study | Number of infants (i/c) | Days of mechanical ventilation | Length of hospital stay (days) | ||
| Intervention | Control | Intervention | Control | ||
| 7/6 | 6.6 ± 2.6 | 7.3 ± 1.7 | 16.4 ± 5.4 | 19.8 ± 2.9 | |
| i = intervention group; c = control group. All variables expressed by mean ± standard deviation. | |||||
|
| Intervention (n = 7) | Control (n = 6) |
| Oxygenation index (0 hours) | 29.8 ± 12.5 (median 25) | 32.4 ± 25 (median 24.3) |
| Oxygenation index (24 hours) | 2.7 ± 2.2 (median 2.8) | 10.4 ± 8.1 (median 9.0) |
| Oxygenation index (48 hours) | 5.0 ± 9.1 (median 1.7) | 5.7 ± 9.1 (median 1.8) |
| AaDO2 (0 hours) | 575.6 ± 91.0 | 589.5 ± 95.8 |
| AaDO2 (24 hours) | 261.0 ± 160.9 | 352.3 ± 177.5 |
| AaDO2 (48 hours) | 178.3 ± 144.0 | 226.0 ± 239.8 |
| i = intervention group; c = control group; AaDO2 = alveolar‐arterial oxygen difference (mmHg), | ||
| Outcome or subgroup title | No. of studies | No. of participants | Statistical method | Effect size |
| 1 Death Show forest plot | 2 | 88 | Risk Ratio (M‐H, Fixed, 95% CI) | 0.42 [0.12, 1.46] |
| 2 Use of ECMO Show forest plot | 2 | 47 | Risk Ratio (M‐H, Fixed, 95% CI) | 0.27 [0.04, 1.86] |
| 3 Death or use of ECMO Show forest plot | 2 | 88 | Risk Ratio (M‐H, Fixed, 95% CI) | 0.33 [0.11, 0.96] |
| 4 Pneumothorax Show forest plot | 2 | 88 | Risk Ratio (M‐H, Fixed, 95% CI) | 0.38 [0.08, 1.90] |
| 5 Oxygenation index Show forest plot | 2 | Mean Difference (IV, Fixed, 95% CI) | Subtotals only | |
| 5.1 measured at 24 hours | 2 | 88 | Mean Difference (IV, Fixed, 95% CI) | ‐1.90 [‐7.56, 3.77] |
| 5.2 measured at 48 hours | 2 | 88 | Mean Difference (IV, Fixed, 95% CI) | ‐6.20 [‐12.11, ‐0.29] |
| 5.3 measured at 72 hours | 2 | 88 | Mean Difference (IV, Fixed, 95% CI) | ‐3.56 [‐8.72, 1.60] |
| 6 Alveolar‐arterial oxygen difference Show forest plot | 1 | Mean Difference (IV, Fixed, 95% CI) | Subtotals only | |
| 6.1 measured at 24 hours | 1 | 66 | Mean Difference (IV, Fixed, 95% CI) | ‐12.0 [‐109.19, 85.19] |
| 6.2 measured at 48 hours | 1 | 66 | Mean Difference (IV, Fixed, 95% CI) | ‐57.0 [‐162.96, 48.96] |
| 6.3 measured at 72 hours | 1 | 66 | Mean Difference (IV, Fixed, 95% CI) | ‐41.0 [‐132.59, 50.59] |
| 7 PaO2/FiO2 Show forest plot | 1 | Mean Difference (IV, Fixed, 95% CI) | Subtotals only | |
| 7.1 measured at 24 hours | 1 | 66 | Mean Difference (IV, Fixed, 95% CI) | ‐1.0 [‐54.61, 52.61] |
| 7.2 measured at 48 hours | 1 | 66 | Mean Difference (IV, Fixed, 95% CI) | 27.0 [‐26.63, 80.63] |
| 7.3 measured at 72 hours | 1 | 66 | Mean Difference (IV, Fixed, 95% CI) | 26.0 [‐24.96, 76.96] |
| Outcome or subgroup title | No. of studies | No. of participants | Statistical method | Effect size |
| 1 Death Show forest plot | 1 | 13 | Risk Ratio (M‐H, Fixed, 95% CI) | 0.18 [0.01, 3.06] |
| 2 Pneumothorax Show forest plot | 1 | 13 | Risk Ratio (M‐H, Fixed, 95% CI) | 0.18 [0.01, 3.06] |
