Study selection
From the study's start to April 26, 2022, a literature search was conducted following PRISMA criteria, resulting in 190 potential articles from sources such as PubMed, Embase, the Cochrane Library, CNKI, and WanFang Data. After removing 29 duplicates using EndNote, 161 articles were selected for further review. 70 articles were analyzed as full-text publications after title and abstract screening, while 65 articles were rejected due to insufficient data. The review process was conducted by two independent researchers (Miao and Niu) and agreement was reached through discussion. The screening process is summarized in the flow chart shown in Fig. 3.
Study design and sample characteristics
The five studies included in the research were published from 2020 to 2022 and originated from five different countries. The sample size ranged from 20 to 30. The main characteristics of the studies are summarized as follows: measuring 24-hour and 48-hour opioid consumption after surgery, evaluating perioperative side effects, and assessing dynamic and static pain scores at various time intervals after block. All studies used ultrasound for localization and only one study used general anesthesia while the rest used spinal anesthesia. Patient-controlled analgesia (PCA) was used postoperatively.
Principal Results
Opioid consumption in 24 hours
Five studies examined the 24-hour postoperative opioid consumption of the PENG block and the FICB and the consumption of various opioids was translated into morphine equivalents administered orally[25]. Since there was heterogeneity, a random effect model was adopted (P = 0. 01, I 2 = 69%). Based on the findings of all of these investigations, it appears that PENG block could reduce the consumption of opioids in 24 hours (MD = -7.26, 95% CI [-14.32, -0.19], P = 0.04, Fig. 4) .
Opioid consumption in 48 hours
A total of three RCTs involving 142 patients showed opioid consumption in 48 hours after hip surgery. The different types of opioids used after surgery were turned into oral morphine equivalents[25]. We utilized a fixed effects model to pool the relevant data because the heterogeneity between the three studies was not significant (I 2 = 34%, P = 0.22). PENG failed to identify a significant association with changes in opioid consumption in 48 hours(MD = -4.74, 95% CI [-12.27, 2.79], P = 0.22, Fig. 5) .
Perioperative Side effects
Four trials with a total of 194 patients provided information on perioperative side effects. Since there was some heterogeneity, we used a random effect model (P = 0. 06, I2 = 59%). Our meta-analysis showed no statistically significant differences existed between the groups (MD = 0.76, 95% CI [0.13,4.57], P = 0.76, Fig. 6) .
Secondary Results
Dynamic pain score at 6 hour after surgery
Three studies, a total of 134 participants reported the NRS at 6 hour after surgery. Because there was no significant heterogeneity (P = 0.37, I 2 = 0%), a fixed effect model was used. Based on the results of all of these studies, we made conclusion that the NRS at 6 hour after surgery of PENG block and FICB block was not significantly different (MD = -0.63, 95% CI [-1.46,0.19], P = 0.13, Fig. 7) .
Static pain score at 6 hour after surgery
Three RCTs have assessed the static pain score at 6 hour after hip surgery. When the fixed effect model was used, there was some heterogeneity, so a random effect model was used (P = 0.14, I 2 = 50%). According to the current meta-analysis, the results of the PENG block had no significant effect compared to the FICB block group (SMD = -0.36, 95% CI [-0.83,0.11], P = 0.13, Fig. 8) .
Dynamic pain score at 24 hour after surgery
Three trials, a total of 134 patients provided data on NRS at 24 hour. According to all three studies, there was no significant difference in NRS scores after 24 hour of block. There is one paper which not only measured the pain levels of the patients between PENG block and FICB using NRS score but also using VAS (VAS data was not extracted) [26]. Compared with the control group, PENG block had no benefit on NRS at 24 hour with moderate heterogeneity (MD = -0.07, 95% CI [-0.86,0.72], P = 0.86; P = 0.08, I 2 = 61%, Fig. 9) .