- Study characteristics
Table 1 listed a total of 9 prospective studies published from 2001 to 2018 were finally included in the present analysis, including 8 RCTs22-29 and 1 prospective case control study30. There were 3 multi-center studies (NCT0264570824, NCT0264313425, one not registered27). There were 8a two-arm studies and 1a threearmstudy29. A total of 1959 subjects were included in the present study, 978 in the physical therapy group and 981 in control group, respectively.
All the physical therapy, we enrolled, was finally concluded into EPVL and PDI. As it demonstrated in the Table 2, 5 studies22-25, 29 focus on EPVL, 129 of which involved tamsulosin. In 4 studies of PDI group, 1 study27 investigated only the mechanical percussion but was classified into a PDI group and the rest of 32, are all PDI, but 128 did not describe the application of diuresis. And LE of all the RCTs was 1b and the score of methodological quality are from 3 to 5. Risk of bias assessment is described in figure 2 (Figure legends - Results of bias assessment.).
- Meta-analysis results
2.1 Stone-Free Rate
A total of 9 eligible studies reported the SFR referred to physical percussion, EPVL and PDI after ESWL, RIRS or not. A higher SFR was provided by physical therapy (OR= 2.72, 95% CI: 1.79–4.14, p = 0.000, figure. 3 (Figure legends - Meta-analysis of stone-free rate and subgroup analysis of different techniques.)). Heterogeneity test showed that I2 = 55% and p = 0.018, which supported random effect used in our meta-analysis.
The funnel plot (figure 4 (Figure legends - Funnel plot of publish bias.)) revealed an symmetry and begg`s test revealed Z = 1.79 and p = 0.074, both of which suggested no presence of a publication bias in our study.
2.2 SFR changed with time
With time after treatment moved on, SFR was also changing. Specifically, the first week (OR: 2.44; 95% CI: 1.00–5.93; p = 0.050), Week 2 (OR: 1.98; 95% CI: 1.34–2.92; p = 0.001), month 1(OR: 3.49; 95% CI: 1.94–6.28; p = 0.000) and month 3(OR: 3.62; 95% CI: 1.71–7.66; p = 0.001, figure 5 (Figure legends - Meta-analysis of stone-free rate in week 1, week 2, month 1 and month 3.))
2.3 First two day stone expulsion
In figure 6 (Figure legends - Meta-analysis of stone expulsion in the first two days.), first two day stone expulsion rate was higher in physical therapy group than control group (OR= 1.58, 95% CI: 1.07–2.32, p = 0.021)
2.4 Complication rate
Overall complication rate was higher in the control group as showed in figure 7 (Figure legends - Meta-analysis of overall complication rate and subgroup analysis of specific complications.), but there was no significant difference (OR: 0.84; 95% CI: 0.61–1.14; p = 0.25). There was also no significance in hematuria (OR: 0.79; 95% CI: 0.42–1.48; p = 0.462), dizziness (OR: 2.87; 95% CI: 0.89–9.32; P = 0.079), lumbago (OR: 0.62; 95% CI: 0.32–1.22; P = 0.164) and urinary infection (OR: 0.75; 95% CI: 0.40–1.41; P= 0.372), respectively.
- Sub-group meta-analysis
3.1 Influence of different physical therapy
As demonstrated in figure 3 (Figure legends - Meta-analysis of stone-free rate and subgroup analysis of different techniques.), when we classified treatment into EPVL and PDI, SFR no matter in EPVL (OR: 2.28; 95% CI: 1.27–4.07; p =0.005) or in PDI (OR: 3.54; 95% CI: 2.22–5.64; p = 0.000) was higher than in the control group.
3.2 Influence of stone location to SFR
In the subgroup analysis (figure 8 (Figure legends - Subgroup analysis of different stone locations in SFR.)), physical intervention group had a higher SFR in LCS (OR: 3.75; 95% CI: 2.39–5.89; p = 0.000). However, there were no significant difference in other groups, UCS (OR: 1.70; 95% CI: 0.14–21.12; p = 0.678), MCS (OR: 5.51; 95% CI: 0.32–94.90; p = 0.240), RPS (OR: 5.08; 95% CI: 0.24–107.94; p =0.297) and UUS (OR: 1.63; 95% CI: 0.97–2.74; p = 0.064) respectively.
3.3 Influence of medical interventions to SFR
According to the analysis result (figure 9 (Figure legends - Subgroup analysis of medical intervention in SFR.)), patients who received no medicines like furosemide and tamsulosin during the physical therapy owned significantly higher SFR (OR: 3.48; 95% CI: 2.48–4.91; p =0.000). However, in tamsulosin group, physical therapy combined tamsulosin or not showed no significant difference compared to tamsulosin group (OR: 1.05; 95% CI: 0.61–1.82; p = 0.855). Similarly, furosemide was also insignificant (OR: 5.69; 95% CI: 0.86–37.78; p = 0.072) to the blank group.
3.4 Influence of drinking water before therapy
When the studies were divided according to whether patients drink enough water before receiving physical therapy as it was showed in Figure 10 (Figure legends - Subgroup analysis of drinking water before physical therapy in SFR.), enough water before therapy was helpful to increase the stone-free rate (OR: 3.31; 95% CI: 2.39–4.60; p = 0.0001). Otherwise, there was no statistical difference (OR: 1.83; 95% CI: 0.57–5.85; p = 0.308) between the studies where no recommends on water before the treatment.