Cirugía de día versus cirugía con estancia nocturna en la colecistectomía laparoscópica
Resumen
Antecedentes
La colecistectomía laparoscópica se utiliza para controlar los cálculos biliares sintomáticos. Existe una controversia considerable sobre si debe hacerse como cirugía de día o como cirugía con estancia nocturna en lo que respecta a la seguridad del paciente.
Objetivos
Evaluar el impacto de la colecistectomía laparoscópica en el día versus la estancia nocturna en resultados orientados al paciente como la mortalidad, los eventos adversos graves y la calidad de vida.
Métodos de búsqueda
Se realizaron búsquedas en el Registro de Ensayos Controlados del Grupo Cochrane Hepatobiliar (Cochrane Hepato‐Biliary Group) y en el Registro Cochrane Central de Ensayos Controlados (Cochrane Central Register of Controlled Trials, CENTRAL) en The Cochrane Library, MEDLINE, EMBASE, Science Citation Index Expanded y mRCT hasta septiembre de 2012.
Criterios de selección
Se incluyeron ensayos clínicos aleatorizados que comparaban la cirugía de día versus la cirugía de estancia nocturna para la colecistectomía laparoscópica, independientemente del idioma o el estado de publicación.
Obtención y análisis de los datos
Dos autores de la revisión evaluaron los ensayos para su inclusión y extrajeron los datos de manera independiente. Los datos se analizaron con el modelo de efectos fijos y con el de efectos aleatorios mediante el análisis de Review Manager 5. Se calcularon los riesgos relativos (RR), la diferencia de medias (DM), o la diferencia de medias estandarizada (DME) con los intervalos de confianza (IC) del 95%, según el análisis por intención de tratar o por datos disponibles.
Resultados principales
Se identificaron un total de seis ensayos con alto riesgo de sesgo que incluían 492 participantes sometidos a colecistectomía laparoscópica en el día (n = 239) versus colecistectomía laparoscópica durante la noche (n = 253) para los cálculos biliares sintomáticos. El número de participantes en cada ensayo osciló entre 28 y 150. La proporción de mujeres en los ensayos varió entre el 74% y el 84%. La media o la mediana de la edad en los ensayos variaron entre 40 y 47 años.
En cuanto a los resultados principales, sólo un ensayo informó sobre la mortalidad a corto plazo. Sin embargo, en el ensayo se afirmó que no hubo muertes en ninguno de los grupos. De los demás resultados se dedujo que no hubo mortalidad a corto plazo en los ensayos restantes. En ninguno de los ensayos se informó sobre la mortalidad a largo plazo. No hubo diferencias significativas en la tasa de eventos adversos graves entre los dos grupos (cuatro ensayos; 391 participantes; 7/191 (tasa ponderada 1,6%) en el grupo de cirugía de día versus 1/200 (0,5%) en el grupo de cirugía de estancia nocturna; cociente de tasas 3,24; IC del 95%: 0,74 a 14,09). No hubo diferencias significativas en la calidad de vida entre los dos grupos (4 ensayos; 333 participantes; DME ‐0,11; IC del 95%: ‐0,33 a 0,10).
No hubo diferencias significativas entre los dos grupos con respecto a las medidas de resultado secundarias de esta revisión: dolor (3 ensayos; 175 participantes; DM 0,02 cm de puntuación en la escala analógica visual; IC del 95%: ‐0,69 a 0,73); tiempo para el retorno a la actividad (2 ensayos, 217 participantes; DM ‐0,55 días; IC del 95%: ‐2,18 a 1,08); y retorno al trabajo (1 ensayo, 74 participantes; DM ‐2,00 días; IC del 95%: ‐10,34 a 6,34). No se observaron diferencias significativas en la tasa de reingreso hospitalario (5 ensayos; 464 participantes; 6/225 (tasa ponderada 0,5%) en el grupo de cirugía de día versus 5/239 (2,1%) en el grupo de cirugía de estancia nocturna (cociente de tasas 1,25; IC del 95%: 0,43 a 3.63) o en la proporción de personas que requieren reingresos hospitalarios (3 ensayos; 290 participantes; 5/136 (proporción ponderada 3,5%) en el grupo de cirugía de día versus 5/154 (3,2%) en el grupo de cirugía con estancia nocturna; RR 1,09; IC del 95%: 0,33 a 3,60). No se observaron diferencias significativas en la proporción de altas fallidas (no se dio de alta según lo previsto) entre los dos grupos (5 ensayos; 419 participantes; 42/205 (proporción ponderada 19,3%) en el grupo de cirugía de día versus 43/214 (20,1%) en el grupo de cirugía de estancia nocturna; RR 0,96; IC del 95%: 0,65 a 1,41). Para todos los resultados, excepto el dolor, la información acumulada fue mucho menor que el tamaño de la información requerida ajustada por la diversidad para excluir los errores aleatorios.
Conclusiones de los autores
La cirugía de día parece tan segura como la cirugía con estancia nocturna en la colecistectomía laparoscópica. La cirugía de día no parece dar lugar a una mejora en ningún resultado orientado al paciente, como el retorno a la actividad normal o la vuelta más temprana al trabajo. Los ensayos clínicos aleatorizados que respaldan estas afirmaciones se ven debilitados por los riesgos de errores sistemáticos (sesgo) y los riesgos de errores aleatorios (juego de azar). Se necesitan más ensayos clínicos aleatorizados para evaluar el impacto de la colecistectomía laparoscópica de la cirugía de día en la calidad de vida así como otros resultados de los pacientes.
Resumen en términos sencillos
Cirugía de día versus cirugía con estancia nocturna en la colecistectomía laparoscópica
Esta revisión compara el alta en el mismo día (cirugía de día) con la estancia nocturna después de la extracción de la vesícula biliar (colecistectomía laparoscópica) para diversas afecciones que afectan la vesícula biliar, pero principalmente para los cálculos biliares que causan dolor.
Los cálculos que se desarrollan en la vesícula biliar pueden causar dolor en la parte superior del abdomen. Este trastorno se trata mediante la extracción quirúrgica de la vesícula biliar a través de una cirugía mínimamente invasiva, un procedimiento que se conoce como colecistectomía laparoscópica. Este procedimiento puede implicar que la persona permanezca en el hospital durante la noche, pero cada vez es más posible realizar la operación y permitir que se vaya a casa el mismo día ("cirugía de día"). Hay cierta controversia sobre si realizar una colecistectomía laparoscópica como cirugía de día es seguro.
Esta revisión tiene por objeto investigar la literatura actual disponible y ofrece una visión general de la evidencia demostrada en recientes ensayos clínicos sobre el tema. Los autores de la revisión identificaron un total de seis ensayos con 492 participantes. Doscientas treinta y nueve personas se sometieron a una colecistectomía laparoscópica planificada como cirugía de día y 253 participantes permanecieron en el hospital durante la noche después del procedimiento. Todos los ensayos tenían un alto riesgo de sesgo (deficiencias metodológicas que podrían hacer posible llegar a conclusiones erróneas al sobrestimar el beneficio o subestimar el daño del procedimiento de cirugía de día o de estancia nocturna). Se examinaron los resultados que se consideran importantes desde la perspectiva del participante y también del proveedor de servicios de salud. Estos resultados incluyen la muerte, complicaciones graves, calidad de vida después del procedimiento, dolor, el tiempo que tardaron las personas en volver a su actividad normal y a su trabajo, reingresos hospitalarios y altas fallidas (no haber sido dados de alta como se había previsto). No hubo una diferencia significativa en la proporción de los que murieron o en la tasa de complicaciones entre el grupo que se sometió a la cirugía de día y los que tuvieron una estancia nocturna. La calidad de vida no difirió significativamente entre los dos grupos. No hubo una diferencia significativa en el tiempo que las personas tardaron en volver a la actividad normal o en volver al trabajo. Tampoco hubo diferencias significativas en las tasas de readmisión en el hospital o de altas fallidas. Los resultados sugieren que la cirugía de día es segura para los pacientes. Es importante señalar que todos los ensayos corrían el riesgo de sesgo y los datos eran escasos, lo que daba lugar a una considerable posibilidad de llegar a conclusiones erróneas debido a errores sistemáticos (sobreestimación de los beneficios o subestimación de los daños de la cirugía de día o la estancia nocturna) y a errores aleatorios (juego de azar). Se necesitan más ensayos aleatorizados para investigar el impacto de la cirugía de día y la estancia nocturna en la calidad de vida y otros resultados de las personas que se someten a una colecistectomía laparoscópica.
Authors' conclusions
Summary of findings
| Day‐surgery versus overnight stay for laparoscopic cholecystectomy | |||||
| Patient or population: patients with laparoscopic cholecystectomy. | |||||
| Outcomes | Illustrative comparative risks* (95% CI) | Relative effect | No of participants | Quality of the evidence | |
| Assumed risk | Corresponding risk | ||||
| Control | Day‐case versus overnight stay for laparoscopic cholecystectomy | ||||
| Short term mortality | See comment | See comment | Not estimable | 86 | ⊕⊕⊝⊝ |
| Serious adverse events | 5 per 1000 | 16 per 1000 | Rate Ratio 3.24 | 391 | ⊕⊝⊝⊝ |
| Quality of life | The mean quality of life in the intervention groups was | SMD ‐0.11 (‐0.33 to 0.1) | 333 | ⊕⊝⊝⊝ | |
| Pain | The mean pain in the intervention groups was | 175 | ⊕⊝⊝⊝ | ||
| Time to return to activity | The mean time to return to activity in the intervention groups was | 217 | ⊕⊝⊝⊝ | ||
| Number of hospital readmissions | 21 per 1000 | 26 per 1000 | Rate ratio 1.25 | 464 | ⊕⊝⊝⊝ |
| Failed discharge | 201 per 1000 | 193 per 1000 | RR 0.96 | 419 | ⊕⊝⊝⊝ |
| *The basis for the assumed risk is the average control group risk across studies. 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 The methods of randomisation were not clear. | |||||
Background
Description of the condition
About 10% to 15% of the adult western population have gallstones (Jørgensen 1987; NIH 1992; Muhrbeck 1995; Halldestam 2004). Between 1% and 4% become symptomatic each year with an acute inflammation of the gallbladder due to obstruction, a condition known as acute cholecystitis (NIH 1992; Halldestam 2004).The gold standard management of symptomatic gallstones is removal of the gallbladder, a procedure known as a cholecystectomy.
In the United States, 1.5 million cholecystectomies are performed annually. In the United Kingdom, approximately 50,000 cholecystectomies are performed annually and the majority of these are performed laparoscopically. Currently, approximately 16,500 of cholecystectomies in the United Kingdom are performed as day‐surgery procedures (HES 2012), with people being discharged the same day as the operation. While it is obvious that performing laparoscopic cholecystectomy can result in cost savings, concerns exist regarding the safety of the patient. The main serious adverse events associated with laparoscopic cholecystectomy are bleeding and bile duct injury (Keulemans 1998; Shamiyeh 2004). The main concern is whether performing day‐surgery can have a negative impact on people because of these serious adverse events because of late recognition of complications. There are also concerns about the adequacy of post‐operative pain control at home.
Description of the intervention
Day‐surgery is defined as patient admission and discharge within the same day, with day‐surgery as the intended management (AAGBI 2011)
How the intervention might work
Reducing post‐operative hospital stay may improve recovery in part by reducing the risk of venous thromboembolism and hospital‐acquired infections (AAGBI 2011).
Why it is important to do this review
Laproscopic cholecystectomy is a major surgical procedure and increasingly this is being conducted in the form of day‐surgery. It is necessary to have up‐to‐date information regarding the safety of laparoscopic cholecystectomy performed as day‐surgery.
This is an update of a previous Cochrane review, last published in 2008 (Gurusamy 2008b).
Objectives
To compare the benefits and harms of day‐surgery versus overnight stay laparoscopic cholecystectomy.
Methods
Criteria for considering studies for this review
Types of studies
We considered only randomised clinical trials for this review, irrespective of language, blinding, or publication status. Quasi‐randomised studies (where the method of allocating participants to a treatment is not strictly random (eg, date of birth, hospital record number, alternation)), cohort studies, and case‐control studies were excluded for benefits because of the bias in their study designs. We considered including quasi‐randomised studies for harms related to day‐surgery.
Types of participants
People undergoing laparoscopic cholecystectomy.
Types of interventions
We included trials comparing day‐surgery laparoscopic cholecystectomy versus overnight stay laparoscopic cholecystectomy.
Types of outcome measures
Primary outcomes
-
Mortality
-
Short‐term mortality (30‐day mortality or in‐hospital mortality).
-
Long‐term mortality (at longest follow‐up).
-
-
Serious adverse events. Adverse events are defined as any untoward medical occurrence not necessarily having a causal relationship with the treatment, but resulting in a dose reduction or discontinuation of treatment (ICH‐GCP 1997). Serious adverse events are defined as any event that would increase mortality, is life‐threatening, requires inpatient hospitalisation, results in a persistent or significant disability, or any important medical event, which might have jeopardised the participant or requires intervention to prevent it.
-
Quality of life.
Secondary outcomes
-
Pain (visual analogue score (VAS) between four and eight hours).
-
Time to return to normal activity.
-
Time to return to work.
-
Hospital readmissions:
-
Total number of hospital readmissions.
-
Number of people requiring hospital admissions.
-
-
Failed discharge.
Search methods for identification of studies
Electronic searches
We searched the Cochrane Hepato‐Biliary Group Controlled Trials Register (Gluud 2012) and the Cochrane Central Register of Controlled Trials (CENTRAL) inThe Cochrane Library, MEDLINE, EMBASE, Science Citation Index Expanded (Royle 2003), and the metaRegister of Controlled Trials (http://www.controlled‐trials.com/mrct/) until September 2012. We have given the search strategies in Appendix 1 with the time spans for the searches.
Searching other resources
We searched the references of the identified trials to identify further relevant trials.
Data collection and analysis
We performed the systematic review following the instructions given in the Cochrane Handbook for Systematic Reviews of Intervention (Higgins 2011) and the Cochrane Hepato‐Biliary Group Module (CHBG module).
Selection of studies
Two authors (JV and KG) identified the trials for inclusion independently of each other. We have listed the excluded studies with the reasons for the exclusion. We resolved any differences through discussion.
Data extraction and management
Two of the review authors (JV and KG) independently extracted the following data:
-
Year and language of publication.
-
Country.
-
Date and duration of the trial.
-
Inclusion and exclusion criteria.
-
Sample size.
-
Post‐randomisation drop‐outs and reasons for these.
-
Participant characteristics such as age and gender.
-
Risk of bias (described below).
-
Outcomes (described above).
We sought any unclear or missing information by contacting the authors of the individual trials. If there was any doubt whether the trials shared the same participants, completely or partially (by identifying common authors and centres), we planned to contact the authors of the trials to clarify whether the trial report had been duplicated. However, we did not find any such trials. We resolved any differences in opinion through discussion until we reached a consensus.
Assessment of risk of bias in included studies
We followed the instructions given in the Cochrane Handbook for Systematic Reviews of Intervention (Higgins 2011) and the Cochrane Hepato‐Biliary Group Module (CHBG module). According to empirical evidence (Schulz 1995; Moher 1998; Kjaergard 2001; Wood 2008; Lundh 2012; Savović 2012a; Savović 2012b), the risk of bias of the trials was assessed based on the following domains: sequence generation, allocation concealment, blinding of participants, personnel, and outcome assessors, incomplete outcome data, selective outcome reporting, and vested interest bias. Risk of bias domains were classified as follows:
Allocation sequence generation
-
Low risk of bias: sequence generation was achieved using computer random number generation or a random number table. Drawing lots, tossing a coin, shuffling cards, and throwing dice are adequate if performed by an independent research assistant not otherwise involved in the trial.
-
Uncertain risk of bias: the method of sequence generation was not specified.
-
High risk of bias: the sequence generation method was not random.
Allocation concealment
-
Low risk of bias: the participant allocations could not have been foreseen in advance of, or during, enrolment. Allocation was controlled by a central and independent randomisation unit. The allocation sequence was unknown to the investigators (eg, if the allocation sequence was hidden in sequentially numbered, opaque, and sealed envelopes).
-
Uncertain risk of bias: the method used to conceal the allocation was not described so that intervention allocations may have been foreseen in advance of, or during, enrolment.
-
High risk of bias: the allocation sequence was likely to be known to the investigators who assigned the participants.
Blinding of participants and personnel
-
Low risk of bias: blinding was performed adequately, or the assessment of outcomes was not likely to be influenced by lack of blinding.
-
Uncertain risk of bias: there was insufficient information to assess whether blinding was likely to introduce bias into the results.
-
High risk of bias: no blinding or incomplete blinding, and the assessment of outcomes was likely to be influenced by lack of blinding.
It is impossible to blind the participants as to whether they received day‐surgery or overnight stay surgery. We rated all the subjective outcomes as being at high risk of bias.
Blinding of outcome assessors
-
Low risk of bias: blinding was performed adequately, or the outcome measurement is not likely to be influenced by lack of blinding.
-
Uncertain risk of bias: there is insufficient information to assess whether the type of blinding used is likely to introduce bias into the estimate of effect.
-
High risk of bias: no blinding or incomplete blinding, and the outcome or the outcome measurement is likely to be influenced by lack of blinding.
Incomplete outcome data
-
Low risk of bias: missing data were unlikely to make treatment effects depart from plausible values. Sufficient methods, such as multiple imputation, have been employed to handle missing data.
-
Uncertain risk of bias: there was insufficient information to assess whether missing data in combination with the method used to handle missing data were likely to introduce bias into the results.
-
High risk of bias: the results were likely to be biased due to missing data.
Selective outcome reporting
-
Low risk of bias: all outcomes were pre‐defined (for example, in a published protocol) and reported, or all clinically relevant and reasonably expected outcomes (mortality and serious adverse events) were reported.
-
Uncertain risk of bias: it is unclear whether all pre‐defined and clinically relevant and reasonably expected outcomes were reported.
-
High risk of bias: one or more clinically relevant and reasonably expected outcomes were not reported, and data on these outcomes were likely to have been recorded.
Vested Interest
-
Low risk of bias: the trial appears to be free of other components that could put it at risk of bias.
-
Uncertain risk of bias: the trial may or may not be free of other components that could put it at risk of bias.
-
High risk of bias: there are other factors in the trial that could put it at risk of bias, eg, for‐profit involvement, authors have conducted trials on the same topic etc.
We considered trials which were classified as being at low risk of bias in all the above domains as trials with a low risk of bias. The remaining trials were classified as trials with a high risk of bias.
Measures of treatment effect
For dichotomous variables, we calculated the risk ratio (RR) with a 95% confidence interval (CI). We reported the risk difference if this was different from the risk ratio (ie, if the risk difference was statistically significant and the risk ratio was not statistically significant, and vice versa). Risk difference takes the 'zero event trials' into account while the risk ratio does not. For continuous variables, we calculated the mean difference (MD) or standardised mean difference (SMD) with a 95% CI. For count data such as serious adverse events, we reported the rate ratio with a 95% CI.
Unit of analysis issues
The unit of analysis was the participant undergoing laparoscopic cholecystectomy.
Dealing with missing data
We performed an intention‐to‐treat analysis whenever possible (Newell 1992). We imputed data for binary outcomes using various scenarios such as 'best‐case' scenario, 'worst‐case' scenario, 'best‐worst case' scenario, and 'worst‐best case' scenario analyses (Gurusamy 2009; Gluud 2012).
For continuous outcomes, we planned to use available‐case analysis. For continuous outcomes, we imputed the standard deviation from P values according to the instructions given in the Cochrane Handbook for Systematic Reviews of Intervention (Higgins 2011) and used the median for the meta‐analysis when the mean was not available. If it was not possible to calculate the standard deviation from a P value or a confidence interval, we imputed the standard deviation as the highest standard deviation noted for that group under that outcome.
Assessment of heterogeneity
We explored heterogeneity by Chi² test with significance set at a P value of 0.10, and measured the quantity of heterogeneity by I² (Higgins 2002).
Assessment of reporting biases
We planned to use visual asymmetry on a funnel plot to explore reporting bias (Egger 1997; Macaskill 2001). We planned to perform the linear regression approach described by Egger 1997 to determine the funnel plot asymmetry. We performed neither of these because fewer than 10 trials were included in this review.
Data synthesis
We performed the meta‐analyses according to the recommendations of The Cochrane Collaboration (Higgins 2011) and the Cochrane Hepato‐Biliary Group Module (CHBG module). We used the software package Review Manager 5 (RevMan 2012). We used a random‐effects model (DerSimonian 1986) and a fixed‐effect model (DeMets 1987). In case of discrepancy between the two models we have reported both results; otherwise we have reported the results of the fixed‐effect model. We summarised the evidence in the 'Summary of findings' table using GRADEpro (GradePro 3.6).
Trial sequential analysis
We used the trial sequential analysis to control for random errors due to sparse data and repetitive testing of the accumulating data for the primary outcomes (CTU 2011; Thorlund 2011). We added the trials according to the year of publication, and if more than one trial was published in a year, we added the trials in alphabetical order according to the last name of the first author. We planned to construct the trial sequential monitoring boundaries on the basis of the diversity‐adjusted required information size (Brok 2008; Wetterslev 2008; Brok 2009; Thorlund 2009, Wetterslev 2009; Thorlund 2010).
We applied trial sequential analysis (CTU 2011; Thorlund 2011) using a diversity‐adjusted required information size calculated from an alpha error of 0.05, a beta error of 0.20, a proportion with the outcome in the control group obtained from the results of the meta‐analysis, and a relative risk reduction of 20% for binary outcomes with two or more trials to determine whether more trials are necessary on this topic. If the trial sequential monitoring boundary or the required information size is reached or the futility zone is crossed, then more trials may not be necessary (Brok 2008; Wetterslev 2008; Brok 2009; Thorlund 2009, Wetterslev 2009; Thorlund 2010). For continuous outcomes, we calculated the required sample size from an alpha error of 0.05, a beta error of 0.20, the variance estimated from the meta‐analysis results of trials with low risk of bias, and a minimal clinically relevant difference of one cm for pain on a visual analogue scale and one day for time to return to activity and planned to calculate the required sample size for time to return to work on the same basis as time to return to activity.
It is not possible to perform trial sequential analysis for count data and for continuous outcomes such as quality of life where standardised mean difference was used as the effect measure.
Subgroup analysis and investigation of heterogeneity
We planned to perform the following subgroup analysis:
-
trials with low risk of bias compared to trials with high risk of bias.
We planned to use the 'test for subgroup differences' to identify the differences between subgroups.
Sensitivity analysis
We planned to perform sensitivity analyses of different methods of imputation for the binary outcomes. We performed a sensitivity analysis excluding the trials in which mean and standard deviation were imputed for the continuous outcomes.
Results
Description of studies
Results of the search
We identified 722 references through the electronic searches of the Cochrane Hepato‐Biliary Group Controlled Trials Register and Cochrane Central Register of Controlled Trials (CENTRAL) in The Cochrane Library (n = 84), MEDLINE (n = 371), EMBASE (n = 116), Science Citation Index Expanded (n = 151), and mRCT (n = 0). We have shown the flow of references in Figure 1. We excluded 242 duplicates and 462 clearly irrelevant references through reading abstracts. We retrieved 18 references for further assessment. We did not find any additional references through scanning reference lists of the identified randomised trials. Of the 18 references, we excluded seven for the reasons listed under the table 'Characteristics of excluded studies'. Eleven publications describing six randomised trials fulfilled the inclusion criteria.
Prisma Flow Chart
Included studies
All six trials were completed and could provide data for the analyses (Keulemans 1998; Hollington 1999; Young 2001; Dirksen 2001; Johansson 2006; Barthelsson 2008). Details of the trials are shown in the table 'Characteristics of included studies'. All the trials assessed day‐surgery versus overnight stay for elective laparoscopic cholecystectomy. Four trials clearly stated that the laparoscopic cholecystectomy was performed for symptomatic gallstones (Keulemans 1998; Dirksen 2001; Johansson 2006; Barthelsson 2008). Two trials did not state the indication for laparoscopic cholecystectomy (Hollington 1999; Young 2001).
A total of 492 participants undergoing elective laparoscopic cholecystectomy were randomised to day‐surgery (n = 239) versus overnight stay surgery (n = 253). The number of participants in each trial ranged from 28 to 150. The proportion of women in the trials varied between 74% and 84%. The mean or median age in the trials varied between 40 and 47 years. Other details have been described in Characteristics of included studies.
Excluded studies
The reasons for exclusion of the studies are listed under the table 'Characteristics of excluded studies'.
Risk of bias in included studies
The risk of bias is summarised in the risk of bias graph (Figure 2) and risk of bias summary (Figure 3). All the trials were at high risk of bias as seen in Figure 3.
Methodological quality graph: review authors' judgments about each methodological quality item presented as percentages across all included studies.
Methodological quality summary: review authors' judgments about each methodological quality item for each included study.
Allocation (selection bias)
Three trials (50%) were considered to have low risk of bias due to the description of random sequence generation (Hollington 1999; Young 2001; Johansson 2006). Four trials (67%) were considered to have low risk of bias due to the description of allocation concealment (Keulemans 1998; Hollington 1999; Young 2001; Johansson 2006).
Blinding (performance bias and detection bias)
None of the trials reported blinding.
Incomplete outcome data (attrition bias)
Two trials (33%) had featured exclusion of participants with acute cholecystitis, acute cholangitis or who postponed surgery because of lack of symptoms. This reflects a real‐life situation and therefore these post‐randomisation drop‐outs would not introduce bias due to missing outcome data (Keulemans 1998; Johansson 2006). One trial (17%) reported no post‐randomisation drop‐outs (Young 2001).
Selective reporting (reporting bias)
Three trials (50%) reported all the primary outcomes and were considered free from bias due to selective reporting (Hollington 1999; Dirksen 2001; Johansson 2006).
Other potential sources of bias
Two trials (33%) demonstrated that there was no vested interest bias (Keulemans 1998; Hollington 1999), while the other four provided no information on this.
Effects of interventions
The main results are summarised in the summary of findings Table for the main comparison.
Mortality
One trial reported short‐term mortality (Dirksen 2001). However, there were no deaths in this trial. Even in the other trials, although short‐term mortality was not reported explicitly, from the type of adverse events that participants suffered, and based on the number of participants included in the other outcomes, we could infer that there was no short‐term mortality in either group. Since there was no mortality in either group, we were unable to use the control group proportion for the calculation of the required information size of the trial sequential analysis. Instead, we used a proportion of 0.2% in the control group based on data from approximately 30,000 patients included in a database in Switzerland (Giger 2011). The proportion of information accrued was only 0.14% of the diversity‐adjusted required information size and so the trial sequential monitoring boundaries were not drawn by the program (Figure 4).
Trial sequential analysis of mortality
The diversity‐adjusted required information size (DARIS) was calculated to 352,564 patients, based on the proportion of patients in the control group with the outcome of 0.2%, a relative risk reduction of 20%, an alpha of 5%, a beta of 20%, and a diversity of 0%. To account for zero event groups, a continuity correction of 0.01 was used in the calculation of the cumulative Z‐curve (blue line). After accruing a total of 492 patients in six trials, only 0.14% of the DARIS has been reached. Accordingly, the trial sequential analysis program does not show the required information size and the trial sequential monitoring boundaries. As shown, the conventional boundaries have also not been crossed by the cumulative Z‐curve.
None of the trials reported long‐term mortality.
Serious adverse events
Four trials reported serious adverse events such as subphrenic collection requiring computerised tomographic scan‐guided drainage, pancreatitis, bleeding requiring intervention, pleuritic chest pain, bile duct injury, and haematoma at port site requiring treatment (Keulemans 1998; Hollington 1999; Dirksen 2001; Johansson 2006). There was no significant difference in the number of serious adverse events between the two groups (rate ratio 3.24; 95% confidence interval (CI) 0.74 to 14.09) (Analysis 1.1). It is not possible to perform trial sequential analysis for rate ratios.
Quality of life
Four trials reported quality of life at one week (Keulemans 1998; Dirksen 2001; Johansson 2006; Barthelsson 2008).There was no significant difference in the quality of life between the two groups (standardised mean difference (SMD) ‐0.11; 95% CI ‐0.33 to 0.10) (Analysis 1.2). It is not possible to perform trial sequential analysis for standardised mean difference data.
Pain
Three trials reported pain (Keulemans 1998; Young 2001; Barthelsson 2008). There was no significant difference in the pain reported between the two groups (mean difference (MD) 0.02 cm visual analogue scale scores; 95% CI ‐0.69 to 0.73) (Analysis 1.3). Trial sequential analysis revealed that the futility zone area was reached suggesting that future trials are unlikely to show clinically significant difference in pain between the two groups (Figure 5).
Trial sequential analysis of pain
Trial sequential analysis of pain showing that the cumulative Z‐curve (blue line) enters the futility area after the third trial. The diversity‐adjusted required information size (DARIS) was 359 participants based on a minimal relevant difference (MIRD) of 1 cm on the visual analogue scale, a variance (VAR) of 11.41, an alpha (a) of 5%, a beta (b) of 20%, and a diversity (D2) of 0%. The results are compatible with lack of difference in pain scores between day‐surgery and overnight stay surgery for laparoscopic cholecystectomy.
Time to return to activity
Two trials reported time taken to return to activity (Hollington 1999; Dirksen 2001). There was no significant difference in the time taken to return to activity between the two groups (MD ‐0.55 days; 95% CI ‐2.18 to 1.08) (Analysis 1.4). Trial sequential analysis revealed that the information was too small (9.21%) to draw the futility area, and the trial sequential boundaries for benefits or harms of day‐surgery were not crossed (Figure 6).
Trial sequential analysis of time to return to activity
The diversity‐adjusted required information size (DARIS) was 2,354 participants based on a minimal relevant difference (MIRD) of 1 day, a variance (VAR) of 74.96, an alpha (a) of 5%, a beta (b) of 20%, and a diversity (D2) of 0%. After accruing 217 participants in two trials, only 9.2% of the DARIS has been reached. Accordingly, the trial sequential analysis program does not show the futility area. Neither the trial sequential boundaries nor the conventional statistical boundaries for benefits or harms of day‐surgery versus overnight stay were crossed by the cumulative Z‐curve (blue line).
Time to return to work
One trial reported time taken to return to work (Keulemans 1998). There was no significant difference in the time taken to return to work between the two groups (MD ‐2.00 days; 95% CI ‐10.34 to 6.34) (Analysis 1.5). We did not perform trial sequential analysis because of the presence of only one trial.
Number of hospital readmissions
Five trials reported the number of hospital readmissions (Keulemans 1998; Hollington 1999; Dirksen 2001; Johansson 2006; Barthelsson 2008). There was no significant difference in the rate of hospital readmissions between the two groups (rate ratio 1.25; 95% CI 0.43 to 3.63) (Analysis 1.6). It is not possible to perform trial sequential analysis for rate ratios.
Number of people requiring readmission
Three trials reported the proportion of people requiring hospital readmissions (Hollington 1999; Dirksen 2001; Barthelsson 2008). There was no significant difference in the proportion of people requiring readmission between the two groups (risk ratio (RR) 1.09; 95% CI 0.33 to 3.60) (Analysis 1.7). The trial sequential analysis revealed that the proportion of information accrued was only 1.38% of the diversity‐adjusted required information size and so the trial sequential monitoring boundaries were not drawn (Figure 7).
Trial sequential analysis of proportion of participants requiring hospital readmission
The diversity‐adjusted required information size (DARIS) was calculated to 21,030 participants, based on the proportion of participants in the control group with the outcome of 3.24%, a relative risk reduction of 20%, an alpha of 5%, a beta of 20%, and a diversity of 0%. To account for zero event groups, a continuity correction of 0.01 was used in the calculation of the cumulative Z‐curve (blue line). After accruing 290 participants in three trials, only 1.38% of the DARIS has been reached. Accordingly, the trial sequential analysis program does not show the required information size and the trial sequential monitoring boundaries. As shown, the conventional boundaries have also not been crossed.
Failed discharge
Five trials reported failed discharge (Keulemans 1998; Hollington 1999; Young 2001; Dirksen 2001; Johansson 2006). There was no significant difference in the proportion of participants who had failed discharge between the two groups (RR 0.96; 95% CI 0.65 to 1.41) (Analysis 1.8). The trial sequential analysis revealed that the proportion of information accrued was only 2.62% of the diversity‐adjusted required information size, and so the trial sequential monitoring boundaries were not drawn (Figure 8).
Trial sequential analysis of proportion of participants with failed discharge
The diversity‐adjusted required information size (DARIS) was calculated to 15,968 participants, based on the proportion of participants in the control group with the outcome of 20.09%, a relative risk reduction of 20%, an alpha of 5%, a beta of 20%, and a diversity of 81.98%. To account for zero event groups, a continuity correction of 0.01 was used in the calculation of the cumulative Z‐curve (blue line). After accruing 419 participants in the five trials, only 2.62% of the DARIS has been reached. Accordingly, the trial sequential analysis program does not show the required information size and the trial sequential monitoring boundaries. As shown, the conventional statistical boundaries have also not been crossed at the end of five trials although the cumulative Z‐curve crossed temporarily the conventional statistical boundary favouring overnight stay after two trials. Such a finding is likely to be a random error.
Sensitivity analysis
In two trials (Keulemans 1998; Hollington 1999), the standard deviation was calculated from the standard error. In the trial by Barthelsson 2008, the standard deviation, the standard error, and the P value were not given, and we used the highest standard deviation in the outcome. Removing the imputed data from each outcome did not cause a change in the results (Analysis 2.1; Analysis 2.2; Analysis 2.3; Analysis 2.4).
Five of the six trials included in this systematic review had post‐randomisation drop‐outs (Keulemans 1998; Hollington 1999; Dirksen 2001; Johansson 2006; Barthelsson 2008). In these groups, we performed a sensitivity analysis for the three binary outcomes (short‐term mortality, number of people requiring hospital readmission, and failed discharge), to investigate best and worst outcomes. It was not possible to impute best and worst scenarios for one trial (Dirksen 2001) as the trial failed to detail the groups to which post‐randomisation drop‐outs belonged. For this reason we did not perform a sensitivity analysis on short‐term mortality, as Dirksen 2001 was the only trial that reported short‐term mortality explicitly. With regards to the number of people requiring hospital readmission, the 'best‐best' scenario results showed no change compared with the original analysis (Analysis 2.5). However, when the outcomes were imputed by the 'worst‐worst' and 'worst‐best' scenarios, the results significantly favoured overnight stay laparoscopic cholecystectomy. In contrast, imputation of outcome according to the 'best‐worst' scenario significantly favoured day‐surgery laparoscopic cholecystectomy (Analysis 2.6; Analysis 2.7; Analysis 2.8). With regards to the number of failed discharge attempts, the imputation of outcomes by the 'best‐best', 'worst‐worst', 'best‐worst', and 'worst‐best' scenarios showed no significant change in the results obtained without imputation (Analysis 2.9; Analysis 2.10; Analysis 2.11; Analysis 2.12).
Variations in meta‐analysis
There was no change in the interpretation of the results by adopting the random‐effects model or by calculating the risk difference rather than the risk ratio.
Subgroup analyses
All the trials were of high risk of bias, so we did not perform any subgroup analysis.
Funnel plot
We did not produce a funnel plot due to the small number of trials (fewer than 10 trials) included in this review.
Discussion
Summary of main results
There were no statistically significant differences in any of the outcomes between day‐surgery and overnight stay laparoscopic cholecystectomy. There was no significant difference in short‐term mortality between the groups of participants undergoing day‐surgery or overnight‐stay elective laparoscopic cholecystectomy. Only one trial reported short‐term mortality explicitly. However, the serious adverse events and the number of participants included in the remaining trials suggest that there was no mortality in the remaining trials. Laparoscopic cholecystectomy is generally considered a low‐morbidity procedure. Long‐term mortality was not reported. However, it is unlikely that long‐term mortality is affected by the decision to admit the patient overnight provided that there was no difference in the short‐term mortality (which was the case in this particular comparison). There was no significant difference in serious adverse events or quality of life between the two groups. There was a suggestion of heterogeneity between the trials with regards to quality of life by the I². Two trials favoured day‐surgery (Dirksen 2001; Johansson 2006) and two trials favoured overnight‐stay surgery (Keulemans 1998; Barthelsson 2008). However, there was significant overlap of confidence intervals and the Chi² test for heterogeneity was not significant. Quality of life is difficult to quantify, and trials report it in markedly different ways. This could be the reason for the heterogeneity noted between the trials.
There was no statistically significant difference in pain, time to return to activity, time to return to work, number of hospital readmissions, number of people requiring hospital readmission, and failed discharge. Inadequate pain control at home or excessive complications because of early discharge would have resulted in an increase in one or more of the above. Considering that none of the above have been affected, it appears that there are no significant differences between day‐surgery and overnight stay laparoscopic cholecystectomy. However, there is still a possibility that these findings are because of lack of evidence of effect and not because of lack of effect. Furthermore, all trials were considered at risk of bias, that is overestimation of benefits and underestimation of harms of day‐surgery or overnight stay. That is why more randomised clinical trials are necessary.
Elective laparoscopic cholecystectomy is a low‐morbidity procedure. Major complications are likely to be noted at the time of laparoscopic cholecystectomy and those with such major complications during operation will not be discharged as day‐surgery patients even if they were admitted to the hospital with an intention to perform the surgery as day‐surgery. The only concern about discharging the people on the same day is that post‐operative complications can be missed. The issue is whether these post‐operative complications would have been identified if the patient had stayed overnight. Many post‐operative complications following laparoscopic cholecystectomy manifest with abdominal pain and so appropriate advice should be given to people to ensure that they return to the hospital if they have abdominal pain uncontrolled by routine analgesia. They should also be advised to return in the presence of dizziness or light headedness (particularly if postural), or exercise intolerance, as these may be early signs of significant bleeding. However, it must be noted that the impact of discharging people on the same day on these rare complications which are not evident within the first four to eight hours of observation in hospital cannot be assessed in randomised clinical trials because of the rare nature of the complications. On the other hand, inadequate pain control can affect the patients' quality of life, return to normal activity, and return to work. Future trials should be powered to measure differences in these outcomes.
We identified four studies that were non‐randomised (Rosen 2001; Burney 2002; Curet 2002; Sharma 2004). These do not report any specific adverse events related to day‐surgery laparoscopic cholecystectomy.
Because of the shorter hospital stay, day‐surgery laparoscopic cholecystectomy is likely to result in lower costs, which will benefit those paying for their surgery and the government in a state‐funded healthcare system.
Overall completeness and applicability of evidence
This review included participants undergoing elective laparoscopic cholecystectomy, which is the predominant management choice for symptomatic gallstones. Four trials clearly stated that the patients underwent elective laparoscopic cholecystectomy for symptomatic gallstones (Keulemans 1998; Dirksen 2001; Johansson 2006; Barthelsson 2008). The indication for elective laparoscopic cholecystectomy was not stated in the remaining two trials (Hollington 1999; Young 2001). However, the findings of this review can be applied to patients undergoing elective laparoscopic cholecystectomy for benign gallbladder polyps and gallbladder dyskinesia since there is no scientific rationale for an interaction between the timing of discharge and the indication for elective laparoscopic cholecystectomy. Hence this review is applicable to most people undergoing elective laparoscopic cholecystectomy. However, the safety of day‐surgery laparoscopic cholecystectomy in acute cholecystitis on a non‐elective basis has not been established and should be investigated further. This review also includes mostly patients at low anaesthetic risk and is applicable only for such people.
Quality of the evidence
All the trials were at a high risk of bias. Blinding of participants, healthcare providers, assessors, and investigators was not reported in any trial. Some trials had a significant proportion of post‐randomisation drop‐outs, introducing missing outcome data bias. This was evident from the sensitivity analysis where imputation of missing data by different scenarios resulted in different results.
While it may be impossible to blind the participants and healthcare providers to the intervention groups, it is possible to blind the outcome assessors and investigators to the intervention groups. Considering that the risk of conversion to open cholecystectomy in elective laparoscopic cholecystectomy is low and it is reasonable to exclude such participants from the analyses as such patients who require conversion to open cholecystectomy will be admitted in real‐life, it is possible to conduct trials with no missing outcome data bias.
Potential biases in the review process
We followed the Cochrane Handbook for Systematic Reviews of Intervention (Higgins 2011) for this review. There were no language, publication status, or sample size restrictions. Thus, we minimised the bias due to selection of trials. However, we have used median for the meta‐analyses when the mean was not available. We have also imputed the standard deviation from P values according to the formulae stated in the Cochrane Handbook for Systematic Reviews of Intervention (Higgins 2011). If the trials stated a P value less than 0.05, we calculated the standard deviation using a P value of 0.05. If the standard deviation could not be calculated because the trial reports just state that there was no statistical significance without mentioning the exact P value, we used the highest standard deviation among the other trials included in the outcome. This imputation of standard deviation may have introduced bias. However, sensitivity analyses performed on imputed values demonstrated no difference in results. The alternative to this imputation is to exclude such trials, but this would make the results even more difficult to interpret.
Agreements and disagreements with other studies or reviews
There are no major changes in the results by following the updated methodology and searches as compared to the previous review version (Gurusamy 2008b).
Prisma Flow Chart
Methodological quality graph: review authors' judgments about each methodological quality item presented as percentages across all included studies.
Methodological quality summary: review authors' judgments about each methodological quality item for each included study.
Trial sequential analysis of mortality
The diversity‐adjusted required information size (DARIS) was calculated to 352,564 patients, based on the proportion of patients in the control group with the outcome of 0.2%, a relative risk reduction of 20%, an alpha of 5%, a beta of 20%, and a diversity of 0%. To account for zero event groups, a continuity correction of 0.01 was used in the calculation of the cumulative Z‐curve (blue line). After accruing a total of 492 patients in six trials, only 0.14% of the DARIS has been reached. Accordingly, the trial sequential analysis program does not show the required information size and the trial sequential monitoring boundaries. As shown, the conventional boundaries have also not been crossed by the cumulative Z‐curve.
Trial sequential analysis of pain
Trial sequential analysis of pain showing that the cumulative Z‐curve (blue line) enters the futility area after the third trial. The diversity‐adjusted required information size (DARIS) was 359 participants based on a minimal relevant difference (MIRD) of 1 cm on the visual analogue scale, a variance (VAR) of 11.41, an alpha (a) of 5%, a beta (b) of 20%, and a diversity (D2) of 0%. The results are compatible with lack of difference in pain scores between day‐surgery and overnight stay surgery for laparoscopic cholecystectomy.
Trial sequential analysis of time to return to activity
The diversity‐adjusted required information size (DARIS) was 2,354 participants based on a minimal relevant difference (MIRD) of 1 day, a variance (VAR) of 74.96, an alpha (a) of 5%, a beta (b) of 20%, and a diversity (D2) of 0%. After accruing 217 participants in two trials, only 9.2% of the DARIS has been reached. Accordingly, the trial sequential analysis program does not show the futility area. Neither the trial sequential boundaries nor the conventional statistical boundaries for benefits or harms of day‐surgery versus overnight stay were crossed by the cumulative Z‐curve (blue line).
Trial sequential analysis of proportion of participants requiring hospital readmission
The diversity‐adjusted required information size (DARIS) was calculated to 21,030 participants, based on the proportion of participants in the control group with the outcome of 3.24%, a relative risk reduction of 20%, an alpha of 5%, a beta of 20%, and a diversity of 0%. To account for zero event groups, a continuity correction of 0.01 was used in the calculation of the cumulative Z‐curve (blue line). After accruing 290 participants in three trials, only 1.38% of the DARIS has been reached. Accordingly, the trial sequential analysis program does not show the required information size and the trial sequential monitoring boundaries. As shown, the conventional boundaries have also not been crossed.
Trial sequential analysis of proportion of participants with failed discharge
The diversity‐adjusted required information size (DARIS) was calculated to 15,968 participants, based on the proportion of participants in the control group with the outcome of 20.09%, a relative risk reduction of 20%, an alpha of 5%, a beta of 20%, and a diversity of 81.98%. To account for zero event groups, a continuity correction of 0.01 was used in the calculation of the cumulative Z‐curve (blue line). After accruing 419 participants in the five trials, only 2.62% of the DARIS has been reached. Accordingly, the trial sequential analysis program does not show the required information size and the trial sequential monitoring boundaries. As shown, the conventional statistical boundaries have also not been crossed at the end of five trials although the cumulative Z‐curve crossed temporarily the conventional statistical boundary favouring overnight stay after two trials. Such a finding is likely to be a random error.
Comparison 1 Day case versus overnight stay for laparoscopic cholecystectomy, Outcome 1 Serious adverse events.
Comparison 1 Day case versus overnight stay for laparoscopic cholecystectomy, Outcome 2 Quality of life.
Comparison 1 Day case versus overnight stay for laparoscopic cholecystectomy, Outcome 3 Pain (cm visual analogue scale score).
Comparison 1 Day case versus overnight stay for laparoscopic cholecystectomy, Outcome 4 Time to return to activity (days).
Comparison 1 Day case versus overnight stay for laparoscopic cholecystectomy, Outcome 5 Time to return to work (days).
Comparison 1 Day case versus overnight stay for laparoscopic cholecystectomy, Outcome 6 Number of hospital readmissions.
Comparison 1 Day case versus overnight stay for laparoscopic cholecystectomy, Outcome 7 Number of people requiring hospital readmission.
Comparison 1 Day case versus overnight stay for laparoscopic cholecystectomy, Outcome 8 Failed discharge.
Comparison 2 Sensitivity analysis, Outcome 1 Quality of life (imputed data removed).
Comparison 2 Sensitivity analysis, Outcome 2 Pain (imputed data removed).
Comparison 2 Sensitivity analysis, Outcome 3 Time to return to activity (imputed data removed).
Comparison 2 Sensitivity analysis, Outcome 4 Number of hospital readmissions (imputed data removed).
Comparison 2 Sensitivity analysis, Outcome 5 Number of people requiring hospital readmission (best‐best).
Comparison 2 Sensitivity analysis, Outcome 6 Number of people requiring hospital readmission (worst‐worst).
Comparison 2 Sensitivity analysis, Outcome 7 Number of people requiring hospital readmission (worst‐best).
Comparison 2 Sensitivity analysis, Outcome 8 Number of people requiring hospital readmission (best‐worst).
Comparison 2 Sensitivity analysis, Outcome 9 Failed discharge (best‐best).
Comparison 2 Sensitivity analysis, Outcome 10 Failed discharge (worst‐worst).
Comparison 2 Sensitivity analysis, Outcome 11 Failed discharge (best‐worst).
Comparison 2 Sensitivity analysis, Outcome 12 Failed discharge (worst‐best).
| Day‐surgery versus overnight stay for laparoscopic cholecystectomy | |||||
| Patient or population: patients with laparoscopic cholecystectomy. | |||||
| Outcomes | Illustrative comparative risks* (95% CI) | Relative effect | No of participants | Quality of the evidence | |
| Assumed risk | Corresponding risk | ||||
| Control | Day‐case versus overnight stay for laparoscopic cholecystectomy | ||||
| Short term mortality | See comment | See comment | Not estimable | 86 | ⊕⊕⊝⊝ |
| Serious adverse events | 5 per 1000 | 16 per 1000 | Rate Ratio 3.24 | 391 | ⊕⊝⊝⊝ |
| Quality of life | The mean quality of life in the intervention groups was | SMD ‐0.11 (‐0.33 to 0.1) | 333 | ⊕⊝⊝⊝ | |
| Pain | The mean pain in the intervention groups was | 175 | ⊕⊝⊝⊝ | ||
| Time to return to activity | The mean time to return to activity in the intervention groups was | 217 | ⊕⊝⊝⊝ | ||
| Number of hospital readmissions | 21 per 1000 | 26 per 1000 | Rate ratio 1.25 | 464 | ⊕⊝⊝⊝ |
| Failed discharge | 201 per 1000 | 193 per 1000 | RR 0.96 | 419 | ⊕⊝⊝⊝ |
| *The basis for the assumed risk is the average control group risk across studies. 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 The methods of randomisation were not clear. | |||||
| Outcome or subgroup title | No. of studies | No. of participants | Statistical method | Effect size |
| 1 Serious adverse events Show forest plot | 4 | Rate Ratio (Fixed, 95% CI) | 3.24 [0.74, 14.09] | |
| 2 Quality of life Show forest plot | 4 | 333 | Std. Mean Difference (IV, Fixed, 95% CI) | ‐0.11 [‐0.33, 0.10] |
| 3 Pain (cm visual analogue scale score) Show forest plot | 3 | 175 | Mean Difference (IV, Fixed, 95% CI) | 0.02 [‐0.69, 0.73] |
| 4 Time to return to activity (days) Show forest plot | 2 | 217 | Mean Difference (IV, Fixed, 95% CI) | ‐0.55 [‐2.18, 1.08] |
| 5 Time to return to work (days) Show forest plot | 1 | 74 | Mean Difference (IV, Fixed, 95% CI) | ‐2.0 [‐10.34, 6.34] |
| 6 Number of hospital readmissions Show forest plot | 5 | Rate Ratio (Fixed, 95% CI) | 1.25 [0.43, 3.63] | |
| 7 Number of people requiring hospital readmission Show forest plot | 3 | 290 | Risk Ratio (IV, Fixed, 95% CI) | 1.09 [0.33, 3.60] |
| 8 Failed discharge Show forest plot | 5 | 419 | Risk Ratio (IV, Fixed, 95% CI) | 0.96 [0.65, 1.41] |
| Outcome or subgroup title | No. of studies | No. of participants | Statistical method | Effect size |
| 1 Quality of life (imputed data removed) Show forest plot | 3 | 259 | Std. Mean Difference (IV, Fixed, 95% CI) | ‐0.19 [‐0.44, 0.05] |
| 2 Pain (imputed data removed) Show forest plot | 1 | 28 | Mean Difference (IV, Fixed, 95% CI) | 0.90 [‐1.09, 2.89] |
| 3 Time to return to activity (imputed data removed) Show forest plot | 1 | 86 | Mean Difference (IV, Fixed, 95% CI) | ‐0.20 [‐3.21, 2.81] |
| 4 Number of hospital readmissions (imputed data removed) Show forest plot | 5 | Rate Ratio (Fixed, 95% CI) | 1.25 [0.43, 3.63] | |
| 5 Number of people requiring hospital readmission (best‐best) Show forest plot | 3 | 336 | Risk Ratio (IV, Fixed, 95% CI) | 1.00 [0.30, 3.32] |
| 6 Number of people requiring hospital readmission (worst‐worst) Show forest plot | 3 | 336 | Risk Ratio (IV, Fixed, 95% CI) | 1.68 [1.03, 2.73] |
| 7 Number of people requiring hospital readmission (worst‐best) Show forest plot | 3 | 336 | Risk Ratio (IV, Fixed, 95% CI) | 4.52 [1.75, 11.70] |
| 8 Number of people requiring hospital readmission (best‐worst) Show forest plot | 3 | 336 | Risk Ratio (IV, Fixed, 95% CI) | 0.29 [0.11, 0.82] |
| 9 Failed discharge (best‐best) Show forest plot | 5 | 451 | Risk Ratio (IV, Fixed, 95% CI) | 0.90 [0.61, 1.33] |
| 10 Failed discharge (worst‐worst) Show forest plot | 5 | 451 | Risk Ratio (IV, Fixed, 95% CI) | 1.09 [0.79, 1.51] |
| 11 Failed discharge (best‐worst) Show forest plot | 5 | 451 | Risk Ratio (IV, Fixed, 95% CI) | 0.75 [0.52, 1.07] |
| 12 Failed discharge (worst‐best) Show forest plot | 5 | 451 | Risk Ratio (IV, Fixed, 95% CI) | 1.25 [0.87, 1.81] |
