Anestesia intravenosa versus anestesia por inhalación para la cirugía ambulatoria pediátrica
Resumen
Antecedentes
La anestesia ambulatoria o de pacientes ambulatorios se realiza en los pacientes que son dados de alta en el mismo día de la cirugía. Las complicaciones perioperatorias, como las náuseas y los vómitos posoperatorios (NVPO), los trastornos conductuales posoperatorios y las complicaciones cardiorrespiratorias deben reducirse al mínimo en la anestesia ambulatoria. La elección de los anestésicos y las técnicas puede influir en la aparición de estas complicaciones, y por lo tanto, retrasar el alta.
Objetivos
El objetivo de esta revisión fue evaluar el riesgo de complicaciones (riesgo de náuseas y vómitos posoperatorios [NVPO], el ingreso o el reingreso al hospital, los trastornos conductuales posoperatorios y las complicaciones respiratorias y cardiovasculares perioperatorias) y los tiempos de recuperación (tiempo hasta el alta de la sala de recuperación y tiempo hasta el alta hospitalaria) al comparar la administración de anestesia intravenosa o por inhalación para la cirugía ambulatoria pediátrica.
Métodos de búsqueda
Se hicieron búsquedas en el Registro Cochrane Central de Ensayos Controlados (Cochrane Central Register of Controlled Trials, CENTRAL) en The Cochrane Library (2013, número 8); MEDLINE (1948 hasta 1 octubre 2013); EMBASE (1974 hasta 1 octubre 2013); Literatura Latino‐Americana e do Caribe em Ciências da Saúde (LILACS) (1982 hasta 1 octubre 2013). También se realizaron búsquedas manuales en revistas relevantes y búsquedas en las listas de referencias de los artículos identificados.
Criterios de selección
Se incluyeron ensayos controlados aleatorios que compararon el uso de anestesia intravenosa versus por inhalación en la cirugía de pacientes ambulatorios pediátricos.
Obtención y análisis de los datos
Dos autores de la revisión evaluaron de forma independiente la calidad de los ensayos y extrajeron los datos. Cuando fue necesario, se solicitó información adicional y aclaraciones de datos publicados de los autores de los ensayos individuales.
Resultados principales
Se incluyeron 16 ensayos que implicaron a 900 niños en esta revisión. La mitad de los estudios no describieron la generación de la secuencia aleatoria, y la mayoría no describió de manera adecuada la ocultación de la secuencia de asignación. Los estudios incluidos revelaron la variabilidad en los tipos y las combinaciones de los fármacos y la duración de la anestesia, lo que limitó el metanálisis y la interpretación de los resultados.
Para la inducción y el mantenimiento de la anestesia hubo una diferencia significativa que favoreció a la anestesia intravenosa con propofol; la incidencia de NVPO fue de 32,6% para el sevoflurano y de 16,1% para el propofol (odds ratio [OR] 2,96; intervalo de confianza [IC] del 95%: 1,35 a 6,49; cuatro estudios, 176 niños, pruebas de baja calidad). El riesgo de trastornos conductuales posoperatorios también favoreció a la anestesia intravenosa, y la incidencia fue de 24,7% para el sevoflurano y 11,5% para el propofol (OR 2,67; IC del 95%: 1,14 a 6,23; cuatro estudios, 176 niños, pruebas de muy baja calidad). No hubo diferencias entre los grupos en el riesgo de complicaciones respiratorias y cardiovasculares intraoperatorias y posoperatorias (OR 0,75; IC del 95%: 0,27 a 2,13; tres estudios, 130 niños, pruebas de muy baja calidad), y no hubo diferencias en el tiempo transcurrido hasta la recuperación de la anestesia y el alta hospitalaria. Estos resultados deben interpretarse con cuidado debido a la heterogeneidad entre los estudios en cuanto al tipo y la duración de las cirugías, los tipos de complicaciones informadas y el alto riesgo de sesgo en casi todos los estudios. Dos estudios (105 participantes) compararon el halotano con el propofol y revelaron la heterogeneidad en la duración de la anestesia y en el tipo de procedimiento ambulatorio. Para el riesgo de NVPO, los resultados de los estudios fueron conflictivos, y para los riesgos de complicaciones intraoperatorias y posoperatorias no hubo diferencias significativas entre los grupos.
Para el mantenimiento de la anestesia hubo una diferencia significativa que favoreció a la anestesia con propofol, con o sin óxido nitroso (N2O), cuando se la comparó con el tiopental y el halotano + N2O (OR 3,23; IC del 95%: 1,49 a 7,02; cuatro estudios, 176 niños, pruebas de baja calidad; y OR 7,44; IC del 95%: 2,60 a 21,26; dos estudios, 87 niños, pruebas de baja calidad), respectivamente. Para el tiempo hasta el alta de la sala de recuperación, no hubo diferencias significativas entre los grupos. Los estudios se realizaron con diferentes cirugías ambulatorias y con un alto riesgo de sesgo.
Cuatro estudios (250 participantes) compararon la inducción de la anestesia por vía intravenosa o por inhalación con la anestesia por inhalación para el mantenimiento, y no encontraron diferencias significativas entre los grupos en ninguno de los resultados (riesgo de NVPO, trastornos conductuales, complicaciones respiratorias y cardiovasculares y el tiempo hasta el alta de la sala de recuperación). El metanálisis no se realizó en esta comparación debido a la significativa heterogeneidad clínica.
No se informó el reingreso al hospital en ninguno de los estudios incluidos. No se informaron otros efectos adversos.
Conclusiones de los autores
Hay pruebas insuficientes para determinar si la anestesia intravenosa con propofol para la inducción y el mantenimiento de la anestesia en los pacientes ambulatorios pediátricos sometidos a cirugía reduce el riesgo de náuseas y vómitos posoperatorios, además del riesgo de trastornos conductuales en comparación con la anestesia por inhalación. Estas pruebas son de baja calidad. Se requieren más estudios de alta calidad para comparar los diferentes tipos de anestesia en diferentes subconjuntos de niños que se someten a la cirugía ambulatoria.
PICO
Resumen en términos sencillos
Anestesia intravenosa versus anestesia por inhalación para los niños sometidos a cirugía ambulatoria.
Pregunta de la revisión
Se examinaron las pruebas acerca de los efectos de la anestesia general (AG) para los niños con cirugía ambulatoria. Los niños recibieron AG por inyección intravenosa (inyección de un fármaco como propofol o tiopental) o por inhalación de un gas como el sevoflurano o el halotano a través de una máscara. Se encontraron 16 estudios que compararon estos dos tipos de AG.
Antecedentes
Se procuró determinar si la anestesia intravenosa o por inhalación redujo el riesgo de resultados, como las náuseas y vómitos posoperatorios y los problemas conductuales en niños sometidos a una cirugía ambulatoria. Las NVPO son uno de los problemas más frecuentes para los niños que se someten a cirugía ambulatoria. Las NVPO constituyen un motivo muy frecuente del retraso en el alta hospitalaria y del ingreso no planificado al hospital de un niño. El dolor posoperatorio puede causar problemas conductuales en los niños. Estos problemas conductuales pueden presentarse como gemidos, inquietud y confusión, además de movimientos físicos como revolcarse en la cama. Los trastornos mentales como las alucinaciones, los delirios y la confusión también pueden ser evidentes cuando un niño despierta de la AG.
Fecha de la búsqueda
Las pruebas se actualizaron hasta el 1 de octubre de 2013.
Características de los estudios
Se incluyeron 16 estudios (900 participantes) en esta revisión. Todos los participantes fueron niños sanos (menores de 15 años de edad) a los que se les programó una cirugía ambulatoria. Los estudios compararon diferentes tipos de fármacos anestésicos intravenosos y por inhalación.
Resultados clave
Los autores encontraron que cuando se comparó la anestesia por inhalación con sevoflurano, la anestesia intravenosa con propofol puede reducir el riesgo de NVPO y el riesgo de problemas de conducta, sin diferencia en el tiempo transcurrido hasta la recuperación de la anestesia y el alta hospitalaria en niños con cirugía ambulatoria. El efecto en las complicaciones fue impreciso. Los estudios que compararon otros agentes anestésicos incluyeron diferentes tipos de procedimientos quirúrgicos, diferentes duraciones del procedimiento y fármacos, lo que dificultó sumar los resultados.
Calidad de la evidencia
La mayoría de los ensayos incluidos en esta revisión fueron de calidad deficiente y tenían alto riesgo de sesgo. Hubo gran variedad en los estudios incluidos de combinaciones de fármacos, la duración de la anestesia y los tipos de procedimientos quirúrgicos, lo que hizo que los resultados fueran inciertos. Se necesitan más ensayos para comparar los tipos de anestesia que se usan en la cirugía ambulatoria pediátrica.
Authors' conclusions
Summary of findings
| Induction and maintenance: sevoflurane + nitrous oxide (N2O) compared to propofol + N2O for paediatric outpatient surgery | ||||||
| Patient or population: paediatric | ||||||
| Outcomes | Illustrative comparative risks* (95% CI) | Relative effect | No of participants | Quality of the evidence | Comments | |
| Assumed risk | Corresponding risk | |||||
| Propofol + N2O | Sevoflurane + N2O | |||||
| Postoperative nausea and vomiting | Study population | OR 2.96 | 176 | ⊕⊕⊝⊝ | ||
| 161 per 1000 | 362 per 1000 | |||||
| Postoperative behavioural disturbances | Study population | OR 2.67 | 176 | ⊕⊝⊝⊝ | ||
| 115 per 1000 | 257 per 1000 | |||||
| Intraoperative and postoperative respiratory and cardiovascular complications | Study population | OR 0.75 | 130 | ⊕⊝⊝⊝ | ||
| 246 per 1000 | 197 per 1000 | |||||
| *The basis for the assumed risk (e.g. the median control group risk across studies) is provided in footnotes. The corresponding risk (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). | ||||||
| GRADE Working Group grades of evidence | ||||||
| 1One study (Picard 2000) evaluated the risk of postoperative nausea and vomiting until discharge from recovery room and three (Baykara 1998; Guard 1998; Gurkan 1999) until 24 hours of postoperative. 8The results are imprecise because the analysis of studies of respiratory complications were based on few patients(90 children) and few events(three). | ||||||
| Maintenance anaesthesia: inhalational compared to intravenous for paediatric outpatient surgery | ||||||
| Patient or population: paediatric | ||||||
| Outcomes | Illustrative comparative risks* (95% CI) | Relative effect | No of participants | Quality of the evidence | Comments | |
| Assumed risk | Corresponding risk | |||||
| Intravenous | Maintenance: inhalational | |||||
| Nausea and vomiting ‐ subgroup analysis: thiopentone or halothane + N2O versus propofol or propofol + N2O | Study population | OR 3.23 | 176 | ⊕⊕⊝⊝ | ||
| 310 per 1000 | 592 per 1000 | |||||
| Nausea and vomiting ‐ subgroup analysis: thiopentone or halothane + N2O versus propofol or propofol + N2O | Study population | OR 7.44 | 87 | ⊕⊕⊝⊝ | ||
| 349 per 1000 | 799 per 1000 | |||||
| *The basis for the assumed risk (e.g. the median control group risk across studies) is provided in footnotes. The corresponding risk (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). | ||||||
| GRADE Working Group grades of evidence | ||||||
| 1 Three studies (Crawford 1998; Hannallah 1994; Reimer 1993) reported these outcome in the recovery and 24hs after surgery and only one study (Borgeat 1990) reported these until discharge from recovery. 4The results are imprecise because the analysis of studies of the risk of PONV were based on few patients (87 children). | ||||||
Background
Description of the condition
The intention of ambulatory or outpatient anaesthesia is for the patient to be admitted and discharged on the same day as their surgical treatment, therapeutic procedure or diagnostic test. The patient is administered general, regional or combined anaesthesia and remains under medical supervision until they have full recovery of their mental and physical functions. They are usually well enough to go home on the same day as their procedure. Improvements in anaesthetic and surgical techniques have resulted in extremely good safety records, leading to a radical increase in the number of surgical procedures performed in ambulatory surgical units (Rabbitts 2010; Troy 2002).
Children make excellent candidates for day case surgery as they are usually healthy, free of systemic disease and typically require straightforward, minor or intermediate surgical procedures. The advantages of outpatient procedures include that the patients return home relatively quickly after their surgery; greater patient and parent comfort; reduced risk of nosocomial infections; the release of hospital beds; greater turnover in surgical centres and decreases in costs to the hospital (Ewah 2006; Troy 2002) and patients.
Description of the intervention
Inhalational and intravenous anaesthesia are two options available today for general anaesthesia in children. Since the first general anaesthetic was administered in 1846, inhalational anaesthetics have played a pivotal role in paediatric anaesthesia. The inhalational anaesthetics that are available for use today in children are sevoflurane, isoflurane, desflurane, enflurane, halothane and nitrous oxide. Intravenous anaesthesia is becoming more widely used due to the advent of improved drugs (rapid‐acting total intravenous anaesthesia (TIVA) drugs), better understanding of pharmacokinetic or pharmacodynamic interactions, simpler and age‐specific delivery systems, non‐pollution of the theatre environment and independence from airway instrumentation (Mani 2010). However, there are some disadvantages such as the need for familiarity with the technique and limited availability of infusion pumps and monitoring devices (Lerman 2009) in some parts of the world.
Intravenous anaesthesia can be performed as a single technique or in combination with either inhalation anaesthetics or regional anaesthesia. The intravenous anaesthetics used in children are ketamine, propofol, thiopental, midazolam, dexmedetomidine and opioids (morphine, fentanyl, sufentanil, alfentanil and remifentanil). General anaesthesia with intravenous drugs only, or TIVA, is gaining increased acceptance because of the availability of short‐acting hypnotics and opioids (Mani 2010; Pandit 1994).
How the intervention might work
Postoperative nausea and vomiting (PONV) is one of the most common complications (its incidence ranges from 8.9% to 42%) affecting patients undergoing ambulatory surgery and is a common reason for delaying hospital discharge and unanticipated hospital admissions (Blacoe 2008; Patel 1988). Prophylactic antiemetics have limited efficacy and they may cause mild side effects, such as sedation or headache, and serious life‐threatening events for instance cardiac arrhythmias (Carlisle 2006; Watcha 2003). In paediatric day surgery, some studies suggest that intravenous anaesthesia is associated with less vomiting in the first 24 hours postoperation than inhalational anaesthetics (Gürkan 1999; Hannallah 1994; Martin 1993; Ved 1996). Most of these studies found that the incidence of early vomiting was significantly different, but not after discharge from the hospital (Gürkan 1999; Hannallah 1994; Ved 1996). Other studies reported no difference in vomiting between anaesthetic techniques (Davis 1997; Picard 2000; Viitanen 1999). The use of nitrous oxide (N2O) during anaesthesia has been associated with increased postoperative vomiting (Tramer 1996; Watcha 2003). The studies have not found any differences in discharge times or admission rates between inhalational and intravenous anaesthesia (Davis 1997; Martin 1993; Pieters 2010; Ved 1996; Viitanen 1999).
The incidence of postoperative pain and behavioural disturbances ranges from 25% to 80%, which can influence duration of stay and discharge from the postanaesthetic care unit (PACU) (Sikich 2004). Pain may increase the incidence of emergence delirium (Lynch 1998) and the behaviour of a child in pain can mimic emergence delirium (Voepel‐Lewis 2003). Some studies suggest that intravenous anaesthetics are associated with less behavioural disturbance than inhalational anaesthesia, while others show conflicting results (König 2009; Pieters 2010; Uezono 2000).
Perioperative respiratory adverse events remain one of the major causes of morbidity and mortality during paediatric anaesthesia (Bhananker 2007; Murat 2004). Maintenance with sevoflurane was associated with an increased incidence of laryngospasm compared with propofol (Von Ungern‐Sternberg 2010). However, propofol significantly increased the risk of bradycardia when compared with other anaesthetics, especially in the group of paediatric patients undergoing potentially arrhythmogenic procedures such as in surgeries for correction of strabismus. The types of anaesthesia induction and maintenance seem to greatly affect the risk of perioperative cardiorespiratory complications (Tramer 1997).
Why it is important to do this review
For day surgery to be an acceptable option, it is necessary for fast emergence and short recovery from anaesthesia with a low incidence of perioperative side effects (Picard 2000). Few children (around 1%) may require inpatient admission (Lonnqvist 2006). The factors found to be associated with duration of stay in the PACU and unplanned admission include postoperative pain, PONV and postoperative behavioural disturbances; these should be minimized (Shnaider 2006).
Both intravenous and inhalational anaesthetic agents are used for induction and maintenance of surgical anaesthesia. For this reason, the choice of an anaesthetic agent and technique has been seen as critical in the avoidance of perioperative complications and can therefore be an important factor in achieving rapid patient discharge (Moore 2002).
Objectives
The objective of this review was to evaluate the risk of complications (the risk of postoperative nausea and vomiting (PONV), admission or readmission to hospital, postoperative behavioural disturbances and perioperative respiratory and cardiovascular complications) and recovery times (time to discharge from recovery ward and time to discharge from hospital) comparing the use of intravenous to inhalational anaesthesia for paediatric outpatient surgery.
Methods
Criteria for considering studies for this review
Types of studies
We included all randomized and quasi‐randomized, published and unpublished controlled clinical studies.
Types of participants
We included outpatients under 18 years old receiving general anaesthesia.
We excluded neonates (defined as postnatal age maximum of 28 days after reaching 40 weeks corrected gestational age).
Types of interventions
-
Inhalational versus intravenous induction
-
Maintenance of anaesthesia with a volatile anaesthetic versus intravenous anaesthetic agent
We excluded participants who received premedication with benzodiazepines and antiemetics.
Types of outcome measures
Primary outcomes
-
The risk of postoperative nausea and vomiting (PONV)
-
The risk of admission or readmission to hospital
-
The risk of postoperative behavioural disturbances (defined as disturbances during the recovery from general anaesthesia that consist of hallucinations, delusions and confusion manifested by moaning, restlessness, involuntary physical activity and thrashing about in the bed)
Secondary outcomes
4. The risk of intraoperative and postoperative respiratory and cardiovascular complications (defined as laryngospasm, bronchospasm, oxygen desaturation to less than 90%, bradycardia, arrhythmia, hypotension or hypertension episodes requiring intervention)
5. Time to discharge from recovery room (time in minutes from discontinuation of anaesthetic to achieving the criteria for discharge from recovery room)
6. Time to discharge from hospital (time in minutes from discontinuation of anaesthetic to discharge)
Search methods for identification of studies
Electronic searches
We searched the Cochrane Central Register of Controlled Trials (CENTRAL) in The Cochrane Library (2013, Issue 8); MEDLINE (1948 to 1 October 2013); EMBASE (1974 to 1 October 2013); Literatura Latino‐Americana e do Caribe em Ciências da Saúde (LILACS) (1982 to 1 October 2013). We searched the literature with both subject headings and free text words. We have provided our MEDLINE search in Appendix 1; LILACS in Appendix 2; EMBASE in Appendix 3 and CENTRAL in Appendix 4.
We searched for ongoing trials in the Current Controlled Trials website (http://www.controlled‐trials.com/).
We did not impose a language restriction. We identified the trials from the sources listed above.
Searching other resources
We scrutinized the reference lists of any identified relevant studies for additional citations.
We contacted specialists in the field, authors of the included trials and pharmaceutical manufacturers for any possible unpublished data.
Data collection and analysis
Selection of studies
Two authors (ACO and EMKS) independently screened the trials identified by the literature search. We documented the reasons for exclusion. See Appendix 5 for the study selection form. We consulted a third author (ANA) if any disagreements arose (at this or at any other stage listed below). We did not include data from the trials in question until a consensus was reached.
Data extraction and management
Two authors (ACO and EMKS) extracted data independently and collected the data on a paper form (see Appendix 6). We cross‐checked the data and resolved any discrepancies by discussion. We used double data entry to prevent data entry errors. We used a standard form to extract the following information:
-
characteristics of the study (design, methods of randomization);
-
participants;
-
interventions and outcomes (types of outcome measures, timing of outcomes, adverse events).
Assessment of risk of bias in included studies
In order to assess the risk of bias we independently assessed the studies included in the review according to the criteria described by Higgins (Higgins 2011a) (see Appendix 7).
We assessed the following domains as 'Yes' (that is low risk of bias), 'Unclear' (uncertain risk of bias) or 'No' (that is high risk of bias).
-
Selection bias: was the sequence generation adequate?
-
Selection bias: was allocation adequately concealed?
-
Detection bias: was there a blinded assessment of outcomes?
-
Attrition bias: were incomplete outcome data adequately addressed?
-
Selective reporting bias: are reports of the study free of the suggestion of selective outcome reporting?
-
Other potential bias: was the study apparently free of other problems that could put it at a high risk of bias?
Measures of treatment effect
For dichotomous variables, we calculated the odds ratio (OR) and 95% confidence interval (CI). For continuous outcomes, we calculated the mean difference (MD) and 95% CI.
Unit of analysis issues
The unit of analysis was based on the individual participant (unit to be randomized for interventions to be compared); that is, the number of observations in the analysis should match the number of individuals randomized.
Dealing with missing data
Irrespective of the type of data, we reported dropout rates in the Characteristics of included studies table and used intention‐to‐treat (ITT) analysis (Higgins 2011b).
Assessment of heterogeneity
We quantified inconsistencies among the pooled estimates using the I2 = ((Q ‐ df)/Q) x 100% statistic, where Q is the Chi2 statistic and df its degrees of freedom. This illustrates the percentage of the variability in effect estimates resulting from heterogeneity rather than sampling error (Higgins 2003; Higgins 2011a).
The thresholds for the interpretation of I2 were as follows:
-
0% to 25% low heterogeneity;
-
25% to 75% moderate; and
-
more than 75% significant heterogeneity (Higgins 2003).
We found significant heterogeneity in some secondary outcomes but there were insufficient data to conduct meta‐analyses by subgroups. In subsequent updates of this review, we will further investigate the possible causes of heterogeneity by exploring the impact of methodological quality and the condition of the individuals (that is patient characteristics, type and duration of surgery, type of disease, nature of the intervention, duration of the intervention, adjuvant drugs). When we found sources of heterogeneity, and if there were sufficient data, we conducted meta‐analyses by subgroups.
When no significant heterogeneity was identified, we computed pooled estimates of the treatment effect for each outcome under a fixed‐effect model. When we detected significant heterogeneity despite subgroups analysis, we calculated the pooled estimate of treatment effects using random‐effects models.
Assessment of reporting biases
We did not evaluate reporting biases or small study effects by drawing a funnel plot (trial effect versus trial size) because there was an insufficient number of studies. In subsequent updates we will assess if a sufficient number of studies (more than 10) are included in the review.
Data synthesis
Qualitative information
We synthesized qualitative information relative to methods, risk of bias, description of participants and outcomes measures and inserted this information in the Characteristics of included studies table.
Quantitative information
For dichotomous variables, we calculated the odds ratio (OR). For continuous variables, we calculated the mean difference (MD). In subsequent updates of this review if continuous data are available for the same aspect but are measured with different instruments (different and not interchangeable units of measure), we will pool these data using the standardized mean difference (SMD). For all statistical methods, we will report the 95% CI when pooling data.
When we identified no significant heterogeneity, we computed pooled estimates of the treatment effect for each outcome using a fixed‐effect model. However, if significant heterogeneity was identified we performed a random‐effects model analysis.
Subgroup analysis and investigation of heterogeneity
We found significant heterogeneity in some secondary outcomes but there were insufficient data to conduct meta‐analyses by subgroups.
In subsequent updates of this review, we will investigate the possible causes of this heterogeneity by further exploring the impact of methodological quality and the condition of the individuals. Thus we will conduct meta‐analyses by subgroups:
-
age and gender of the children;
-
different inhalation and intravenous anaesthetics;
-
use of nitrous oxide;
-
use of loco‐regional anaesthesia;
-
use of antiemetics;
-
type and duration of surgery (tonsillectomy and adenoidectomy, strabismus and others)
Sensitivity analysis
In subsequent updates of this review, once the number of available studies are adequate, we will perform a sensitivity analysis to explore the causes of heterogeneity and the robustness of the results. We will include the following factors in the sensitivity analysis, separating studies according to: allocation concealment quality (adequate or unclear or inadequate); blinding outcome assessment (adequate or unclear or inadequate or not performed); rate of withdrawals for each outcome; length of follow‐up; patient characteristics; type of intervention; type of outcome.
Summary of findings
We used the principles of the GRADE system (Guyatt 2008) to assess the quality of the body of evidence associated with specific outcomes (the risk of PONV, the risk of admission or readmission to hospital, the risk of postoperative behavioural disturbances, the risk of intraoperative and postoperative respiratory and cardiovascular complications) in our review and we constructed a summary of findings (SoF) table using the GRADE software. The GRADE approach appraises the quality of a body of evidence based on the extent to which one can be confident that an estimate of effect or association reflects the item being assessed. The assessment of the quality of a body of evidence considers within study risk of bias (methodological quality), the directness of the evidence, heterogeneity of the data, precision of effect estimates and risk of publication bias.
Results
Description of studies
See Characteristics of included studies; Characteristics of excluded studies; Characteristics of studies awaiting classification.
Results of the search
See Figure 1.
Study flow diagram.
We identified 917 citations from the database searches. After screening by title and abstract we obtained full paper copies for 35 citations that were potentially eligible for inclusion in the review. Of these 35 citations, 19 did not fulfil our inclusion criteria and were excluded for the reasons described in the Characteristics of excluded studies. Therefore, we included 16 studies in this review (see Characteristics of included studies).
Two studies (Gouda 2002; Moussa 1995) were found to be incomplete for the assessment of risks of bias and data extraction. We are waiting for the authors to release the full papers (see Characteristics of studies awaiting classification). We did not find any ongoing and unpublished studies.
Included studies
We included 16 studies (Annila 1999; Baykara 1998; Borgeat 1990; Cohen 2003; Cohen 2004; Crawford 1998; Glaisyer 2005; Guard 1998; Gurkan 1999; Hannallah 1994; Madan 2001; Picard 2000; Reimer 1993; Viitanen 1998; Viitanen 1999; Weir 1993) in this review. Of those 16 studies, one study was in Turkish (Baykara 1998). The 16 studies enrolled a total of 900 participants. Two articles analysed the same sample, reporting different outcomes in each publication (Cohen 2003; Cohen 2004).
One study used a cross‐over design (Glaisyer 2005) whereby two interventions were studied in the same group of patients. The order of the intervention was randomized. All other studies claimed to be randomized controlled trials (RCTs).
Types of participants
All participants were healthy children (aged under 15 years), American Society of Anesthesiology (ASA) grades one and two (ASA I and II), who were scheduled for elective outpatient surgery.
Six studies involved elective ear, nose and throat (ENT) procedures (Annila 1999; Baykara 1998; Borgeat 1990; Picard 2000; Viitanen 1998; Viitanen 1999); four ophthalmic procedures, three of these were strabismus surgery (Gurkan 1999; Reimer 1993; Weir 1993) and one a non‐invasive diagnostic ophthalmic procedure (Madan 2001). In addition, six other studies (Cohen 2003; Cohen 2004; Crawford 1998; Glaisyer 2005; Guard 1998; Hannallah 1994) involved other ambulatory surgical procedures (urological, general surgery, dental, orthopaedic, plastic surgery).
Types of interventions
Eight studies compared sevoflurane anaesthesia to propofol (Baykara 1998; Cohen 2003; Cohen 2004; Glaisyer 2005; Guard 1998; Gurkan 1999; Picard 2000; Viitanen 1999) while three studies compared halothane anaesthesia to propofol (Annila 1999; Hannallah 1994; Madan 2001). Four studies compared the induction of anaesthesia with thiopental and maintenance with halothane to propofol anaesthesia (Borgeat 1990; Crawford 1998; Hannallah 1994; Reimer 1993). One study compared the induction of anaesthesia with thiopental to halothane (Viitanen 1998) and another study compared the maintenance of anaesthesia with halothane to propofol (Weir 1993).
Types of outcome measures
Ten studies reported the risk of PONV (Borgeat 1990; Cohen 2004; Glaisyer 2005; Gurkan 1999; Hannallah 1994; Madan 2001; Picard 2000; Reimer 1993; Viitanen 1999; Weir 1993) and four studies reported the incidence of vomiting alone (Baykara 1998; Cohen 2003; Guard 1998; Viitanen 1998). Nine studies reported the incidence of postoperative nausea or vomiting, or both, until 24 hours after surgery (Baykara 1998; Cohen 2004; Crawford 1998; Guard 1998; Gurkan 1999; Hannallah 1994; Reimer 1993; Viitanen 1999; Weir 1993) and one reported it until discharge from recovery (Borgeat 1990).
Eight studies reported the risk of behavioural disturbances in the recovery room (Baykara 1998; Borgeat 1990; Cohen 2003; Glaisyer 2005; Gurkan 1999; Picard 2000; Viitanen 1998; Viitanen 1999). The risk of admission or readmission to hospital was not reported in any study.
The time to discharge from the recovery room was reported in 12 studies (Baykara 1998; Borgeat 1990; Cohen 2004; Crawford 1998; Glaisyer 2005; Guard 1998; Hannallah 1994; Madan 2001; Picard 2000; Reimer 1993; Viitanen 1998; Viitanen 1999) and time to discharge from hospital was reported in two studies (Baykara 1998; Guard 1998).
The intraoperative and postoperative respiratory and cardiovascular complications that were reported are listed in the table Characteristics of included studies.
Excluded studies
We excluded 19 studies for the reasons described in the table Characteristics of excluded studies. We excluded seven studies because the patients were premedicated with benzodiazepines. The use of premedication can change the following outcomes: time to emergence, the incidence of nausea or vomiting, or both, and the incidence of behavioural disturbances.
Three trials (Elliott 2002; Elliott 2003; Moore 2003) were potentially based on the same subset of patients.
Risk of bias in included studies
Risk of bias graph: review authors' judgements about each risk of bias item presented as percentages across all included studies.
Risk of bias summary: review authors' judgements about each risk of bias item for each included study.
Allocation
In eight studies, the procedure for randomization was performed using a computer‐based random listing or a table of random numbers and in eight studies the methods used for randomization were unclear. See Characteristics of included studies.
Most studies (12) probably did not perform an adequate method of allocation sequence concealment (see Figure 2).
Blinding
In three trials, outcome assessment was probably not blinded (Baykara 1998; Borgeat 1990; Viitanen 1998). In most trials the outcomes were assessed by an investigator who was blind to the type of anaesthesia used (Annila 1999; Cohen 2003; Cohen 2004; Crawford 1998; Glaisyer 2005; Guard 1998; Gurkan 1999; Hannallah 1994; Madan 2001; Picard 2000; Reimer 1993; Viitanen 1999; Weir 1993).
Incomplete outcome data
In nine studies follow‐up was 100% and the analysis included all participants with outcome data. Outcome data were incomplete in the other studies.
Selective reporting
In three studies (Baykara 1998; Gurkan 1999; Weir 1993) some specific outcomes were not well described.
In two other studies (Cohen 2003; Viitanen 1998) the pre‐specified outcomes were not reported in the results (see the details in Characteristics of included studies).
Other potential sources of bias
The outcomes of the included studies were analysed in different periods: one study evaluated the risk of PONV until discharge from the recovery room (Borgeat 1990) and nine until 24 hours postoperation. Three studies reported the risk of postoperative behavioural disturbances until discharge from the recovery room (Baykara 1998; Gurkan 1999; Picard 2000) and one during emergence from anaesthesia (Guard 1998). The studies used different scales to define this outcome.
The number of included studies was also insufficient to evaluate reporting biases and small study effects by constructing a funnel plot.
Effects of interventions
See: Summary of findings for the main comparison Induction and maintenance: sevoflurane + nitrous oxide (N2O) compared to propofol + N2O for paediatric outpatient surgery; Summary of findings 2 Maintenance anaesthesia: inhalational compared to intravenous for paediatric outpatient surgery
Induction and maintenance: inhalational versus intravenous
In this comparison we have grouped the studies that used induction and maintenance of anaesthesia with the same intravenous or inhaled drugs in both periods.
1. Sevoflurane versus propofol
Four studies (176 participants) were included in the comparison of sevoflurane versus propofol with all patients receiving nitrous oxide (Baykara 1998; Guard 1998; Gurkan 1999; Picard 2000). Among the studies, different surgeries were performed with the duration of anaesthesia ranging from 50 to 88 minutes. Three studies (Guard 1998; Gurkan 1999; Picard 2000) were classified as at high risk of bias for allocation concealment and one study (Baykara 1998) was classified as unclear for method of randomization and high risk for blinding of outcome assessment. A significant difference was observed favouring the intravenous anaesthesia in the risk of presenting with nausea and vomiting postoperatively, OR of 2.96 (95% CI 1.35 to 6.49) (Analysis 1.1). However, Baykara 1998 and Picard 2000 studied children undergoing tonsillectomy and in Gurkan 1999 the children underwent strabismus surgery. These types of operations are associated with an increased risk of vomiting in children independent of the type of anaesthetic used.
For postoperative behavioural disturbances, the meta‐analysis of these four studies showed a significant difference with less risk in the group receiving intravenous anaesthesia (OR 2.67; 95% CI 1.14 to 6.23) (Analysis 1.2). However, three of these studies (Baykara 1998; Guard 1998; Gurkan 1999) did not describe their definition of behavioural disturbances.
For intraoperative and postoperative respiratory and cardiovascular complications, three studies (130 participants) reported this outcome (Baykara 1998; Guard 1998; Gurkan 1999) and there were no differences between the groups (Analysis 1.3). Two studies (Baykara 1998; Guard 1998) reported respiratory complications (laryngospasm) in the recovery room and another study (Gurkan 1999) reported bradycardia during the induction of anaesthesia. The meta‐analysis of the studies showed significant heterogeneity (49%).
In three studies (136 participants) the time to discharge from the recovery room was reported (Baykara 1998; Guard 1998; Picard 2000) and there was no significant difference between the groups (MD 3.61; 95% CI 0.32 to 6.90) (Analysis 1.4). For time to discharge from hospital, only two studies (83 participants) reported this outcome (Baykara 1998; Guard 1998) and there was no significant difference between the groups, however, this analysis showed moderate heterogeneity (I2 = 74%) (Analysis 1.5). The risk of admission or readmission to hospital was not reported in the studies.
2. Halothane versus propofol
Two studies (105 participants) that compared halothane to propofol during induction and maintenance of anaesthesia showed heterogeneity in the duration of anaesthesia (maximum time of eight minutes in Madan 2001 and 88 minutes in Hannallah 1994), and the type of procedure was different between the studies. Besides that, in Madan 2001 only the halothane group received nitrous oxide while in Hannallah 1994 both groups did. For the risk of PONV, one study did not report any cases of nausea and vomiting in either group (Madan 2001) while Hannallah 1994 reported a significant decrease of this outcome in the group receiving intravenous anaesthesia with propofol (OR 22.15; 95% CI 2.58 to 189.95). For intraoperative and postoperative respiratory and cardiovascular complications, the two studies analysed this outcome and there were no significant differences between the groups. Conflicting results were observed between the two studies in the time to discharge from the recovery room. In Madan 2001 the group receiving inhalational anaesthesia had a better outcome in contrast to Hannallah 1994 in which a better outcome was observed in the intravenous group. The time to discharge from hospital and the risk of postoperative behavioural disturbances were not reported in these studies.
Maintenance of anaesthesia: inhalational versus intravenous
Seven studies were included in this analysis (Borgeat 1990; Cohen 2004; Crawford 1998; Glaisyer 2005; Hannallah 1994; Reimer 1993; Weir 1993). The included studies analysed different drug combinations making it difficult to sum the results.
Four studies (176 participants) compared thiopentone and halothane + N2O versus propofol and propofol + N2O (Borgeat 1990; Crawford 1998; Hannallah 1994; Reimer 1993) for different ambulatory surgeries with the duration varying from 36 to 75 minutes. Regarding the risk of PONV, a significant difference was observed favouring the group receiving intravenous anaesthesia with propofol (OR 3.23; 95% CI 1.49 to 7.02) (Analysis 2.1). Nevertheless, Borgeat 1990 and Reimer 1993 studied children undergoing otorhinolaryngology and strabismus surgery, respectively, and in two other studies different ambulatory surgeries were performed. Moreover, in three studies (Borgeat 1990; Hannallah 1994; Reimer 1993) the method of randomization was unclear and all four studies were classified as high risk of bias for the allocation concealment. Three of these studies (125 participants) reported the time to discharge from the recovery room (Borgeat 1990; Crawford 1998; Reimer 1993) and there were no significant differences between groups, with significant heterogeneity in the analysis (Analysis 2.2). The criteria used for recovery room discharge were different. No other outcomes were reported in these studies.
Two studies (87 participants) compared thiopentone and halothane + N2O versus propofol and propofol (Crawford 1998; Reimer 1993) for ambulatory surgeries lasting less than 60 minutes. The risk of PONV favoured the group receiving intravenous anaesthesia with propofol (OR 7.44; 95% CI 2.60 to 21.26) (Analysis 2.1). There were no significant differences between groups (Analysis 2.2) for the time to discharge from the recovery room. However, the studies were classified as at high risk of bias for the concealment of allocation and randomization.
Weir 1993 (78 participants) compared halothane + N2O and halothane + N2O versus halothane + N2O and propofol + N2O and there was significantly less PONV in the intravenous group (OR 2.91; 95% CI 1.15 to 7.37), the only outcome reported in the study.
Glaisyer 2005 (21 participants) compared propofol and sevoflurane + N2O versus propofol and propofol in painful procedures lasting less than five minutes in a cross‐over trial, and a significant difference was observed favouring intravenous anaesthesia in the time to discharge from recovery (MD 18.60; 95% CI 8.35 to 28.85) in an analysis of the two periods (42 observations). There was no episode of PONV or difference in the risk of postoperative behavioural disturbances between the groups.
Cohen 2004 (56 participants) compared sevoflurane + N2O and sevoflurane + N2O versus sevoflurane + N2O and propofol + N2O for ambulatory surgical procedures, and no significant difference between groups was observed in the risk of PONV and in the time to discharge from the recovery room. In Cohen 2003, which probably analysed the same sample as in Cohen 2004, there was no difference in the risk of postoperative behavioural disturbances between the groups.
Induction of anaesthesia: inhalational versus intravenous
Only four studies (250 participants) compared the induction of anaesthesia by inhalational (halothane or sevoflurane) or intravenous (thiopentone or propofol) administration, with the maintenance phase being equal in both groups; all the participants received inhalation anaesthesia with halothane or sevoflurane for maintenance (Annila 1999; Hannallah 1994; Viitanen 1998; Viitanen 1999). The included studies analysed different drug combinations and the duration of the surgeries ranged from 22 minutes to 68 minutes, therefore a meta‐analysis was not performed. There were no significant differences between groups in all outcomes (PONV, behavioural disturbances, respiratory and cardiovascular complications and time to discharge from recovery room).
Discussion
Summary of main results
There were 16 studies comparing the outcomes following paediatric outpatient surgery between intravenous and inhalational anaesthesia. There was significant clinical heterogeneity between the included studies, such as types and combinations of drugs, the nature of the surgical procedure and the duration of the anaesthesia, limiting the usefulness of the results.
Fourteen studies (697 children) compared intravenous anaesthesia to inhalational anaesthesia. Six of the studies were for ENT procedures, four were for ophthalmic surgery and four were for other procedures. The risk of postoperative nausea and vomiting (PONV) was considered within a period of 24 hours because the included studies looked at this outcome at different periods of time. The use of intravenous anaesthesia (propofol) for both induction and maintenance of anaesthesia decreased the risk of PONV compared to the use of inhalational anaesthesia (sevoflurane) (OR 2.96; 95% CI 1.35 to 6.49, four studies, 176 children) (summary of findings Table for the main comparison). The risk of PONV also decreased for the induction and maintenance of anaesthesia with propofol (+ N2O) versus induction with thiopental and maintenance with halothane and N2O (OR 3.23; 95% CI 3.28 to 23.44, four studies, 176 children); and induction and maintenance with propofol versus induction with thiopentone and maintenance with halothane plus N2O (OR 7.44; 95% CI 2.60 to 21.26, two studies, 87 children) (summary of findings Table 2). However, there was heterogeneity between studies in the type and duration of the operations and there were significant problems with the quality of these randomized trials (summary of findings Table for the main comparison; summary of findings Table 2). The risk of admission or readmission to hospital was not reported in these studies.
The selected studies used different scales and most of them did not describe the postoperative behavioural disturbances criteria. The use of propofol reduced the risk of postoperative agitation compared to inhalational anaesthesia with sevoflurane (OR 2.67; 95% CI 1.14 to 6.23, four studies, 176 children) (summary of findings Table for the main comparison). There were no differences between the sevoflurane and propofol groups in the risk of intraoperative and postoperative respiratory and cardiovascular complications (summary of findings Table for the main comparison). The types of reported complications were not the same among the studies and they were analysed at different times in the perioperative period. There was no difference between groups for time to discharge from the recovery room and hospital. The different studies used different discharge scoring criteria to complete patient assessments and ensure patient readiness for discharge that were predetermined by institutional standing orders and the duration and nature of the surgery.
So far the evidence does not allow us to recommend the use of propofol alone in the induction or maintenance of anaesthesia in relation to these outcomes.
Overall completeness and applicability of evidence
The results of this review show that the use of propofol may be associated with a reduction in the risk of PONV within the first 24 hour period when compared to sevoflurane or mixed anaesthesia (thiopental with halothane), and a reduction of postoperative behavioural disturbances when compared to sevoflurane; these results were based on studies with high risk of bias and significant clinical heterogeneity, such as different durations and types of procedures and the use of different scales for behavioural assessment.
Quality of the evidence
The drugs used, how they were combined and the duration of the anaesthesia varied greatly in the included studies, which made it difficult to provide a meaningful analysis. Besides, there was no study classified as providing high or moderate quality evidence for the specific outcomes (the risk of PONV and postoperative behavioural disturbances). Most studies probably did not conceal allocation, half of them did not describe the generation of a randomized sequence, and three studies probably did not have blinded outcome assessment. Thus, the results of this review are limited and should be considered with caution.
Potential biases in the review process
We attempted to minimize bias in a number of ways; two review authors assessed eligibility for inclusion, carried out data extraction and assessed risk of bias. Each worked independently. The non‐surrogate endpoint in outpatient settings is the risk of admission or readmission to hospital caused by postoperative complications (dehydration, vomiting, agitation, pain), but no studies in this review reported this outcome. The risk of PONV, the risk of postoperative behavioural disturbances, and the risk of intraoperative and postoperative respiratory and cardiovascular complications were probably used as surrogate endpoints. There are a number of reasons why children vomit besides the type of anaesthetic agent, such as type and duration of surgery and the use of opioids. Only six studies in this review used opioids in the perioperative period. In an attempt to reduce bias, this review excluded studies in which participants received antiemetics and premedication with a benzodiazepine because they reduce postoperative vomiting and behavioural changes.
Agreements and disagreements with other studies or reviews
One meta‐analysis (Sneyd 1998) reported on paediatric patients and adults undergoing the following types of surgery under general anaesthesia: arthroscopy or minor orthopaedic; breast; eye; facial; oral or dental; ear, nose and throat; gynaecological laparoscopy, other gynaecological; squint; and other or unspecified surgery. The objective of this review was to investigate the incidence of PONV following maintenance of anaesthesia with propofol compared with inhalational agents. This meta‐analysis has not been updated since. Thirteen paediatric studies were included. The conclusion of this review (Sneyd 1998) was that patients who received maintenance of anaesthesia with propofol had a significantly lower incidence of PONV when compared with inhalational agents. This lower incidence was regardless of induction agent, the choice of inhalational agent, the presence or absence of nitrous oxide (N2O), the patient's age, or use of opiates.
Study flow diagram.
Risk of bias graph: review authors' judgements about each risk of bias item presented as percentages across all included studies.
Risk of bias summary: review authors' judgements about each risk of bias item for each included study.
Comparison 1 Induction and maintenance: sevoflurane + nitrous oxide (N2O) versus propofol + N2O, Outcome 1 Nausea and vomiting.
Comparison 1 Induction and maintenance: sevoflurane + nitrous oxide (N2O) versus propofol + N2O, Outcome 2 Postoperative behavioural disturbances.
Comparison 1 Induction and maintenance: sevoflurane + nitrous oxide (N2O) versus propofol + N2O, Outcome 3 intraoperative and postoperative respiratory and cardiovascular complications.
Comparison 1 Induction and maintenance: sevoflurane + nitrous oxide (N2O) versus propofol + N2O, Outcome 4 Time to discharge recovery (min).
Comparison 1 Induction and maintenance: sevoflurane + nitrous oxide (N2O) versus propofol + N2O, Outcome 5 Time to discharge from hospital (min).
Comparison 2 Maintenance: inhalational versus intravenous, Outcome 1 Nausea and vomiting.
Comparison 2 Maintenance: inhalational versus intravenous, Outcome 2 Time to discharge from recovery (min).
| Induction and maintenance: sevoflurane + nitrous oxide (N2O) compared to propofol + N2O for paediatric outpatient surgery | ||||||
| Patient or population: paediatric | ||||||
| Outcomes | Illustrative comparative risks* (95% CI) | Relative effect | No of participants | Quality of the evidence | Comments | |
| Assumed risk | Corresponding risk | |||||
| Propofol + N2O | Sevoflurane + N2O | |||||
| Postoperative nausea and vomiting | Study population | OR 2.96 | 176 | ⊕⊕⊝⊝ | ||
| 161 per 1000 | 362 per 1000 | |||||
| Postoperative behavioural disturbances | Study population | OR 2.67 | 176 | ⊕⊝⊝⊝ | ||
| 115 per 1000 | 257 per 1000 | |||||
| Intraoperative and postoperative respiratory and cardiovascular complications | Study population | OR 0.75 | 130 | ⊕⊝⊝⊝ | ||
| 246 per 1000 | 197 per 1000 | |||||
| *The basis for the assumed risk (e.g. the median control group risk across studies) is provided in footnotes. The corresponding risk (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). | ||||||
| GRADE Working Group grades of evidence | ||||||
| 1One study (Picard 2000) evaluated the risk of postoperative nausea and vomiting until discharge from recovery room and three (Baykara 1998; Guard 1998; Gurkan 1999) until 24 hours of postoperative. 8The results are imprecise because the analysis of studies of respiratory complications were based on few patients(90 children) and few events(three). | ||||||
| Maintenance anaesthesia: inhalational compared to intravenous for paediatric outpatient surgery | ||||||
| Patient or population: paediatric | ||||||
| Outcomes | Illustrative comparative risks* (95% CI) | Relative effect | No of participants | Quality of the evidence | Comments | |
| Assumed risk | Corresponding risk | |||||
| Intravenous | Maintenance: inhalational | |||||
| Nausea and vomiting ‐ subgroup analysis: thiopentone or halothane + N2O versus propofol or propofol + N2O | Study population | OR 3.23 | 176 | ⊕⊕⊝⊝ | ||
| 310 per 1000 | 592 per 1000 | |||||
| Nausea and vomiting ‐ subgroup analysis: thiopentone or halothane + N2O versus propofol or propofol + N2O | Study population | OR 7.44 | 87 | ⊕⊕⊝⊝ | ||
| 349 per 1000 | 799 per 1000 | |||||
| *The basis for the assumed risk (e.g. the median control group risk across studies) is provided in footnotes. The corresponding risk (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). | ||||||
| GRADE Working Group grades of evidence | ||||||
| 1 Three studies (Crawford 1998; Hannallah 1994; Reimer 1993) reported these outcome in the recovery and 24hs after surgery and only one study (Borgeat 1990) reported these until discharge from recovery. 4The results are imprecise because the analysis of studies of the risk of PONV were based on few patients (87 children). | ||||||
| Outcome or subgroup title | No. of studies | No. of participants | Statistical method | Effect size |
| 1 Nausea and vomiting Show forest plot | 4 | 176 | Odds Ratio (M‐H, Fixed, 95% CI) | 2.96 [1.35, 6.49] |
| 2 Postoperative behavioural disturbances Show forest plot | 4 | 176 | Odds Ratio (M‐H, Fixed, 95% CI) | 2.67 [1.14, 6.23] |
| 3 intraoperative and postoperative respiratory and cardiovascular complications Show forest plot | 3 | 130 | Odds Ratio (M‐H, Fixed, 95% CI) | 0.75 [0.27, 2.13] |
| 4 Time to discharge recovery (min) Show forest plot | 3 | 136 | Mean Difference (IV, Fixed, 95% CI) | 3.61 [0.32, 6.90] |
| 5 Time to discharge from hospital (min) Show forest plot | 2 | 83 | Mean Difference (IV, Random, 95% CI) | 19.54 [‐10.51, 49.59] |
| Outcome or subgroup title | No. of studies | No. of participants | Statistical method | Effect size |
| 1 Nausea and vomiting Show forest plot | 4 | Odds Ratio (M‐H, Fixed, 95% CI) | Subtotals only | |
| 1.1 Thiopentone\Halothane+N2O vs Propofol\Propofol+N2O | 4 | 176 | Odds Ratio (M‐H, Fixed, 95% CI) | 3.23 [1.49, 7.02] |
| 1.2 Thiopentone\Halothane+N2O vs Propofol\Propofol | 2 | 87 | Odds Ratio (M‐H, Fixed, 95% CI) | 7.44 [2.60, 21.26] |
| 2 Time to discharge from recovery (min) Show forest plot | 3 | Mean Difference (IV, Random, 95% CI) | Subtotals only | |
| 2.1 Thiopentone\Halothane + N2O vsPropofol\Propofol + N2O | 3 | 125 | Mean Difference (IV, Random, 95% CI) | 11.07 [‐3.83, 25.97] |
| 2.2 Thiopentone\Halothane +N2O vs Propofol\Propofol | 2 | 87 | Mean Difference (IV, Random, 95% CI) | ‐1.61 [‐10.37, 7.16] |
