After a systematic search of the English and Chinese databases, 614 publications were included in the full-text review, and in the end, a total of 23 [9,10,11, 15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34] publications met the inclusion criteria of this study. The literature screening process is shown in Fig. 1. All studies were published during 2002–2021; 7 studies were from China, and 18 studies were retrospective. Moreover, the 23 studies involved 2844 patients in total, of whom 1430 patients were treated with PSP and 1414 patients were treated with SP. In addition, the results of the quality evaluation of the literature showed that most of the studies had a low risk of bias with a mean NOS score of 7.04 (median: 7); more information on the basic characteristics of the literature is shown in Table 1.

Flow chart of literature screening

Twenty studies reported on the outcome of transient facial palsy occurring in the study subjects after surgery, in 1,292 patients who were treated with PSP and in 1,346 patients who were treated with SP. The results of the heterogeneity evaluation showed good homogeneity between the included studies (I2 = 27%, P = 0.13), and the combined effect size was calculated using a fixed effects model. The meta-analysis results showed that patients had a lower risk of developing transient facial palsy after treatment with PSP compared to SP treatment modalities (OR = 0.33; 95% CI 0.26–0.41), as shown in Fig. 2.

The effect of PSP and SP on patients with transient facial palsy

Eighteen studies reported on the outcomes of permanent facial palsy occurring in the study subjects after surgery, of which four studies did not identify any patients who developed permanent facial palsy during the follow-up [11, 20, 26, 29]. Thus, only the remaining 14 studies were meta-analysed. The results of the heterogeneity evaluation showed no heterogeneity between the included studies (I2 = 0%, P = 0.70), and the combined effect size was calculated using a fixed effects model. The results of the meta-analysis showed that the PSP treatment modality reduced the occurrence of permanent facial palsy in patients after surgery (OR = 0.28; 95% CI 0.16–0.52), as shown in Fig. 3.

The effect of PSP and SP on permanent facial palsy in patients

Twenty studies reported on the outcome of Frey syndrome occurring in the study subjects after surgery, with a total of 1,145 treated with PSP and a total of 1,272 treated with SP experiencing the syndrome. The results of the heterogeneity evaluation showed heterogeneity among the included studies (I2 = 52%, P = 0.004), and the combined effect size was calculated using a random effects model. The meta-analysis results showed that patients treated with PSP had a lower risk of developing Frey syndrome postoperatively compared to those treated with SP (OR = 0.36; 95% CI 0.23–0.56), as shown in Fig. 4. Sensitivity was analysed by presenting literature data one by one. When one study was excluded [29], the heterogeneity was reduced to 38%, and the combined effect size was 0.38 (95% CI 0.29–0.50) using a fixed effects model for the meta-analysis.

The effect of PSP and SP on Frey syndrome in patients

Six studies reported on the outcome of salivary fistulas occurring in the study subjects after surgery. The results of the heterogeneity evaluation showed good homogeneity between the included studies (I2 = 1%, P = 0.41), and the meta-analysis was performed using a fixed effects model. The results of this study showed no statistically significant difference in PSP versus SP in the development of postoperative salivary fistula (OR = 0.70; 95% CI 0.40–1.24), as shown in Fig. 5.

The effect of PSP and SP on patients with salivary fistula

Fifteen studies reported on the outcome of tumour recurrence in patients after surgery, of which 10 studies did not find tumour recurrence during follow-up [10, 20,21,22,23,24, 26, 28, 29, 34]. Therefore, only the remaining five studies were meta-analysed in this study. The results of the heterogeneity evaluation showed no heterogeneity among the included studies (I2 = 0%, P = 0.97), and the meta-analysis was performed using a fixed effects model. The final combined effect size showed that the effect of PSP and SP on postoperative tumour recurrence rate was not statistically significant (OR = 1.41; 95% CI 0.48–4.20), as shown in Fig. 6.

The effect of PSP and SP on the tumor recurrence rate of patients

Eight studies reported on the outcome of postoperative sialocele in the patients. Only one of them did not detect sialocele during follow-up [26], and the remaining seven studies were meta-analysed in this study. The results of the heterogeneity evaluation showed (I2 = 43%, P = 0.11) good homogeneity among the included studies, and the combined effect size was calculated using a fixed effects model. The results of the meta-analysis showed that the effect of PSP and SP on the occurrence of sialocele in patients after surgery did not show a statistically significant difference (OR = 1.48; 95% CI 0.78–2.83), as shown in Fig. 7.

The effect of PSP and SP on sialocele in patients

Five studies reported on the occurrence of postoperative haematomas in patients. The results of the heterogeneity evaluation showed no heterogeneity between the included studies (I2 = 0%, P = 0.58), and the meta-analysis was performed using a fixed effects model. The final combined effect size showed that the PSP treatment modality reduced the occurrence of postoperative haematoma in patients compared to SP (OR = 0.34), but this positive effect was not statistically significant (95% CI 0.11–1.01), as shown in Fig. 8.

The effect of PSP and SP on the hematoma of patients

Ten studies reported on the operative time for PSP and SP. The results of the heterogeneity evaluation showed heterogeneity between the included studies (I2 = 97%, P < 0.00001), and the combined effect size was calculated using a random effects model. The results of the meta-analysis showed that the operative time of PSP was significantly lower than that of SP (MD: − 27.35; 95% CI − 39.66, − 15.04), as shown in Fig. 9. Upon excluding studies one by one for the sensitivity analysis, no significant sources of heterogeneity were found, indicating relatively stable heterogeneity between the included studies.

Comparison results of PSP and SP surgery time (min)