Gasless laparoscopy versus conventional laparoscopy and laparotomy: A systematic review on the safety and efficiency

Gasless laparoscopy (GL) emerged to overcome the clinical and financial challenges of pneumoperitoneum and is often seen as a viable option for use in resource‐limited settings as a means of saving costs and resources. This study aims to systematically review the evidence available on the safety and efficiency of GL compared with conventional laparoscopy (CL) and laparotomy.


| INTRODUCTION
Around 5 billion people worldwide lack access to safe, timely and affordable surgical, obstetrics and anaesthesia care. 1 Because of lack of data, it becomes difficult to estimate the burden of abdominal surgical conditions. 2 It is worth noting that a Global Burden of Disease study estimates the global mortality rate due to appendicitis as 0.7, biliary and liver conditions as 1.6 and intestinal obstruction as 3.9 per 100 000 population. 3 A study in India showed that 1.1% of deaths (0-69 years) are attributed to acute abdominal conditions (72 000 deaths in 2020). 4 In high-income settings, these conditions are managed using diagnostic imaging or diagnostic laparoscopy. The increased utilisation of laparoscopy over laparotomy stems from several benefits such as reducing pain, reducing wound infection, promoting faster post-operative recovery and return to quality of life. [5][6][7][8][9] However, laparoscopic abdominal surgery is resource intensive, requiring expensive surgical equipment, extensive training, general anaesthesia and a ventilator to safely manage the abdominal insufflation.
In lower-income settings, conventional laparoscopy (CL) is often unaffordable due to expensive air-tight trocars, limited availability of carbon dioxide gas and lack of general anaesthesiologists needed for patients undergoing pneumoperitoneum and ventilator support. 10 Patients may instead undergo laparotomy or go without surgery.
Hence, many conditions are untreated, or undiagnosed because of the lack of diagnostic equipment and procedures. This leads to increased morbidity and mortality that could have been avoided by prompt access to resources.
In the early 1990s gasless laparoscopy (GL) was introduced in a small scale primarily in resource-limited settings, as a means to avoid the financial and clinical challenges associated with pneumoperitoneum and general anaesthesia. 11 The pneumoperitoneum is avoided by mechanical elevation of the anterior abdominal wall using a planar lifting system, inflatable balloon or other means to allow an overview of the abdominal organs. Several methods such as gynaecological, [12][13][14] upper or lower gastrointestinal [15][16][17][18][19][20] and exploratory diagnostic procedures 21 have been performed through single or multiple incisions using a gasless technique.
The technique has been described and evaluated in several small studies and safety outcomes such as peri-operative complication rates [22][23][24] and efficiency outcomes 19,25 have been compared with conventional intervention methods.
GL has been proposed to have a variety of potential benefits.
One key benefit is the lower cost compared with CL due to less expensive equipment, maintenance and anaesthesia needs [24][25][26] as it can be used under spinal or local anaesthesia. 17,27,28 It could thereby act as a diagnostic option in resource-limited settings. Some studies have reported GL to be associated with shorter duration of hospital stay for inpatients. 26,29 Others suggest that GL decreases the haemodynamic burden due to abdominal insufflation in patients with haemodynamic compromise (e.g., exploratory laparoscopy for abdominal trauma). 17,27,28 Other suggested benefits are improved safety outcomes in pregnancy, reduced post-site metastasis rates and reduced contamination in some procedures. [30][31][32]

| Study selection
The inclusion criteria for the included studies followed the population, interventions, comparator, outcomes (PICO) tool, where the population of interest was adults aged 18 years or older without any restriction to race, gender or geography (Table 2). Studies included all procedures that were done in a hospital from 1990 and onwards published in English. All excluded studies were tracked to understand how many studies were excluded based on language (Table 3). The intervention used in this study is GL. Gasless procedures were limited to general surgical procedures or conditions, or any trial involving diagnostic laparoscopy performed for general surgical investigation. It includes any variation of gasless technology and device design, any variation of gasless technique and any variation of a gasless training program. The comparison groups were general surgery patients who had CL, laparotomy or received no interventions. All study designs were included; such as case-control, cohort and randomised controlled trials (RCTs). The main outcomes that this study investigated were the safety and efficiency of GL.

| Data extraction and analysis
Eight reviewers participated in the overall review process, with two independent reviewers reviewing each article, and a third senior reviewer involved to discuss discrepancies. Studies were initially screened based on titles and abstracts followed by full-text screening against the inclusion criteria. Data were extracted into Micro-

| RESULTS
The search yielded 1080 studies from all the database searches. As shown in Figure 1  reported as an indication for either GL, CL or OL or both (CL and OL).
Because of the heterogeneity of data, a meta-analysis was deemed non-feasible in this systematic review, however descriptive statistics were used to pool data for cholecystectomies due to relative homogeneity. For all other procedures only a qualitative analysis was performed. Data extracted from all 43 included articles are shown in Table 4. 15,16,19,28,29,33, The main outcomes of the study examined both the efficiency and safety of GL in comparison to other interventions. Efficiency is reported as the mean setup time (minutes) of the patient for the operation, mean duration of the operation (minutes) and the total length of stay in the hospital (days) for those undergoing cholecystectomies across the 24 RCTs.

| Efficiency
In patients undergoing cholecystectomies, the mean setup time was 13.14 min (95% CI À0. 16    from RCT studies reporting a total of 440 patients ( Figure 3B).
From non-RCT studies, such as case-controls and cohorts (10 stud-  Figure 4B).

| Safety
The safety of the surgical techniques has been structured into three  Readmissions were reported equal in five studies for those who underwent GL and five studies for those who underwent CL. Only one study reported all-cause mortality in GL ( Figure 5).

| Risk-of-bias assessment
More than 50% of the RCTs reported low risk of bias as shown in Figure 6A, except in allocation concealment, which showed that around 35% of studies had a low risk of bias, and per-study risk-of-bias analysis  is shown in Figure 6B. Around 75% of non-randomised studies had a low risk of bias except around 35% of studies that showed a low risk of confounding and selection bias, as shown in Figure 7A, and the perstudy risk-of-bias analysis is shown in Figure 7B. and this needs to be confirmed in the future. 75   F I G U R E 6 (A) Cochrane risk-of-bias graph-assessment of randomised studies. Risk of bias in randomised controlled trials. Bias 1: outcomesdetection bias; bias 2: participants-performance bias; bias 3: incomplete outcome (attrition bias); bias 4: selective outcomes (reporting bias); bias 5: allocation concealment; bias 6: random sequence generation. (B) Cochrane risk-of-bias summary of all randomised studies.

| CONCLUSION
GL seems to be a promising alternative to CL or OL in a variety of procedures. However, the lack of homogeneity of data precluded a comprehensive meta-analysis in this systematic review. Hence, the

FUNDING INFORMATION
Open access funding provided through the University of Gothenburg.

CONFLICT OF INTEREST STATEMENT
The authors declare that they have no competing interests.
F I G U R E 7 (A) Cochrane risk-of-bias graph-assessment of non-randomised studies. Non-randomised controlled trials: bias 1: intervention bias; bias 2: measurement outcome bias; bias 3: missing data bias; bias 4: reporting bias; bias 5: risk of confounding bias; bias 6: selection bias. (B) Cochrane risk-of-bias summary of all non-randomised studies.