| Figure 1. Study flow diagram. |
| Figure 2. Risk of bias graph: review authors' judgements about each risk of bias item presented as percentages across all included studies. |
| Figure 3. Risk of bias summary: review authors' judgements about each risk of bias item for each included study. |
| Figure 4. Trial sequential analysis of mortality
The diversity-adjusted required information size (DARIS) was calculated to 352,564 participants, based on the proportion of participants 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 634 participants in seven trials, only 0.18% of the DARIS has been reached. Accordingly, the trial sequential analysis does not show the required information size and the trial sequential monitoring boundaries. As shown, the conventional boundaries (dotted red line) have also not been crossed by the cumulative Z-curve. |
| Figure 5. Trial sequential analysis of serious adverse events
The diversity-adjusted required information size (DARIS) was calculated to 234,831 participants, based on the proportion of participants in the control group with the outcome of 0.3%, 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 634 participants in seven trials, only 0.27% of the DARIS has been reached. Accordingly, the trial sequential analysis does not show the required information size and the trial sequential monitoring boundaries. As shown, the conventional boundaries (dotted red line) have also not been crossed by the cumulative Z-curve. |
| Figure 6. Trial sequential analysis of conversion to open cholecystectomy
The diversity-adjusted required information size (DARIS) was calculated to 40,918 participants, based on the proportion of patients in the control group with the outcome of 1.7%, 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 581 participants in five trials, only 1.42% of the DARIS has been reached. Accordingly, the trial sequential analysis does not show the required information size and the trial sequential monitoring boundaries. As shown, the conventional boundaries (dotted red line) have also not been crossed by the cumulative Z-curve. |
| Figure 7. Trial sequential analysis of operating time
The diversity-adjusted required information size (DARIS) was 1124 participants based on a minimal relevant difference (MIRD) of 15 minutes, a variance (VAR) of 385.03, an alpha (a) of 5%, a beta (b) of 20%, and a diversity (D^{2}) of 95.2%.The cumulative Z-curve (blue line) crosses the conventional boundary (dotted red line) and the trial sequential monitoring boundaries (continuous red line) after the seventh trial. Although the DARIS has not been reached after accrual of 855 participants in nine trials, the results suggest that operating time is longer with fewer-than-four-ports laparoscopic cholecystectomy as compared with four-ports laparoscopic cholecystectomy with low risk of random errors. |
| Figure 8. Trial sequential analysis of hospital stayThe diversity-adjusted required information size (DARIS) was 222 participants based on a minimal relevant difference (MIRD) of 1 day, a variance (VAR) of 1.29, an alpha (a) of 5%, a beta (b) of 20%, and a diversity (D^{2}) of 81.55%. Seven hundred and thirty-one participants were accrued in six trials. The cumulative Z-curve (blue line) crosses the DARIS after the third trial (vertical red line). However, it does not cross the conventional boundaries (dotted red line). This suggests that there is no significant difference in the length of hospital stay between fewer-than-four-ports laparoscopic cholecystectomy versus four-ports laparoscopic cholecystectomy with low risk of random errors. |
| Figure 9. Trial sequential analysis of hospital stay with smallest confidence interval used as minimal relevant differenceThe diversity-adjusted required information size (DARIS) was 2802 participants based on a minimal relevant difference (MIRD) of 0.28 days, a variance (VAR) of 1.29, an alpha (a) of 5%, a beta (b) of 20%, and a diversity (D^{2}) of 81.55%. After accruing 731 participants in six trials, only 26.09% of the DARIS has been reached. Accordingly, the trial sequential analysis does not show the futility area. As shown, the cumulative Z-curve (blue line) does not cross the trial sequential boundaries (red line). Conventional boundaries (dotted red line) have also not been crossed by the cumulative Z-curve. |
| Figure 10. Trial sequential analysis of return to normal activity The diversity-adjusted required information size (DARIS) was 200 participants based on a minimal relevant difference (MIRD) of 1 day, a variance (VAR) of 6.35, an alpha (a) of 5%, a beta (b) of 20%, and a diversity (D^{2}) of 0%. Three hundred and twenty-five participants were accrued in two trials. The cumulative Z-curve crosses the trial sequential monitoring boundaries (continuous red lines) and the conventional boundaries (dotted red line) after the second trial. The results are compatible with significantly fewer days to return to normal activity in the fewer-than-four-ports group with four-ports group with low risk of random errors. |
| Figure 11. Trial sequential analysis of return to normal activity with smallest confidence interval used as minimal relevant difference
The diversity-adjusted required information size (DARIS) was 304 participants based on a minimal relevant difference (MIRD) of 0.81 day, a variance (VAR) of 6.35, an alpha (a) of 5%, a beta (b) of 20%, and a diversity (D^{2}) of 0%. Three hundred and twenty-five participants were accrued in two trials. The cumulative Z-curve crosses the trial sequential monitoring boundaries (continuous red lines) and the conventional boundaries (dotted red line) after the second trial. The results are compatible with significantly fewer days to return to normal activity in the fewer-than-four-ports group with four-ports group with low risk of random errors. |
| Analysis 1.1. Comparison 1 Fewer-than-four ports versus four ports, Outcome 1 Serious adverse events. |
| Analysis 1.2. Comparison 1 Fewer-than-four ports versus four ports, Outcome 2 Quality of life. |
| Analysis 1.3. Comparison 1 Fewer-than-four ports versus four ports, Outcome 3 Conversion to open cholecystectomy. |
| Analysis 1.4. Comparison 1 Fewer-than-four ports versus four ports, Outcome 4 Operating time. |
| Analysis 1.5. Comparison 1 Fewer-than-four ports versus four ports, Outcome 5 Hospital stay. |
| Analysis 1.6. Comparison 1 Fewer-than-four ports versus four ports, Outcome 6 Proportion discharged as day-surgery. |
| Analysis 1.7. Comparison 1 Fewer-than-four ports versus four ports, Outcome 7 Return to normal activity. |
| Analysis 1.8. Comparison 1 Fewer-than-four ports versus four ports, Outcome 8 Return to work. |
| Analysis 1.9. Comparison 1 Fewer-than-four ports versus four ports, Outcome 9 Cosmetic score. |
| Analysis 2.1. Comparison 2 Fewer-than-four ports versus four ports (sensitivity analysis), Outcome 1 Quality of life. |
| Analysis 2.2. Comparison 2 Fewer-than-four ports versus four ports (sensitivity analysis), Outcome 2 Operating time. |
| Analysis 2.3. Comparison 2 Fewer-than-four ports versus four ports (sensitivity analysis), Outcome 3 Hospital stay. |
| Analysis 2.4. Comparison 2 Fewer-than-four ports versus four ports (sensitivity analysis), Outcome 4 Return to normal activity. |