Study characteristics
We initially identified 666 potential articles from Pubmed and Embase. After duplicated records excluded, 67 studies left then we Screened the titles, abstracts and topics. The full-texts were evaluated by the excluding criteria, a final 17 studies15-31 embracing 714 cases and 801 controls were included for extracting data. The literature search and selection process as the flow chart is shown in Fig. 1. We show the included studies in Table 1, these studies were published from 2014 to 2022. Among them, 11 studies were retrospectively designed and the others were prospectively designed. All studies were based on Asia, Europe, and USA. The baseline data including publication year, Study period, Country, age, body mass index (BMI), operation mode, sample size. Also we have exhibited the final study quality scores based on Jadad scale and NOS scoring rules in Table 1.
Operative time
All the including studies have reported the results of the operative time in 3D (n=714) and non-3D group (n=801). A random effect model was used because the heterogeneity among studies is high (I2>50%). The final meta-analysis showed the operative time of Patients undergoing PN with 3D assisted was shorter than Patients undergoing PN without 3D (SMD -0.24, 95% CI –0.59 to –0.09; p < 0.001)( Fig. 2).
Ischemic time
There were 14 studies15,17-19,21-22,24-31 have reported the results of Ischemic time in 3D (n=633) and non-3D group (n=727). The heterogeneity among studies is high (I2 >50%), then a random effect model was used. The final meta-analysis showed the ischemic time of Patients undergoing PN with 3D assisted was shorter than Patients undergoing PN without 3D (SMD -0.42, 95% CI –0.64 to –0.21; p < 0.001) ( Fig. 3).
Estimated blood loss
There were 15 studies15,17-24,26-31 have showed the results of estimated blood loss in 3D (n=669) and non-3D group (n=756). The heterogeneity among studies is high (I2 >50%), then a random effect model was used. The result showed there was no significant difference of estimated blood loss of patients undergoing PN with or without 3D assisted (SMD -0.16, 95% CI –0.37 to 0.05; p < 0.001) (Fig. 4).
Decline in eGFR
There were 8 studies15,17-18,20,22,24,28,30 have showed the results of declining of eGFR in 3D (n=366) and non-3D group (n=433). The eGFR declining in patients undergoing PN with 3D or without 3D assisted was showed no significant difference (SMD -0.05, 95% CI –0.34 to 0.23; p < 0.001) (Fig. 5).
Length of hospital stay
The results of length of hospital stay including in 8 studies17, 18, 23, 26-30. The final meta-analysis showed there was no significant difference of hospital stay length between 3D (n=441) and non-3D group (n=496) (SMD -0.03, 95% CI –0.82 to 0.77; p < 0.001). A random effect model was used because the heterogeneity among studies is high (I2 >50%) (Fig. 6).
Opening collecting system
Four studies compared PN with 3D (n = 169) to without 3D (n = 231) in the case opening collecting system17,22-23,31. The heterogeneity among studies is low (I2 <50%), then a fixed effect model was used. The likelihood of opening collecting system was lower for PN with 3D (OR 0.26, 95% CI 0.14–0.47; p = 0.391) (Fig. 7).
Positive surgical margins and intraoperative complication
We extracted data on positive surgical margins from 8 studies17,19,22,25-26,29-31 (n(3D)=416,n(non-3D)=488) and intraoperative complications from 10 studies(17-18,22,24-26,28-31) (n(3D)=478,n(non-3D)=545). Fixed effect models were used based on the heterogeneity test. The final results indicated that there was no significant difference in positive surgical margin rate (OR 0.56, 95% CI 0.25–1.23; p = 0.958) and intraoperative complication rate (OR 0.75, 95% CI 0.45–1.26; p = 0.958) (Fig. 8.9).