A Meta-Analysis and Systematic Review of Holmium Laser Treatment of Bladder Stones

Results

Literature Retrieve Results

Initially, a total of 432 references were identified, among which 112 studies were from PubMed, 58 were from Embase, 42 were from Medline, 51 were from Springer, 28 were from Ovid, and 141 were from Web of Sciences. Next, NoteExpress3.2 was used to exclude 222 duplicates. Subsequently, the 2 experts read the title and the abstract of each article to exclude those that obviously did not meet the inclusion criteria. After this step, 72 studies were left. After the 2 experts read the full text of each of the 72 articles, 10 studies remained, which were included in the meta-analysis. All articles described RCTs that had been published between 2000–2020. In total, the 12 studies comprised 1,642 research subjects^[3–14] (see Figure 1 and Table 1).

The Bias Risk Assessment

The assessment items included: (I) the generation of the random sequence: 10 of the included studies stated that the subjects were grouped based on different surgery methods, which indicated a low risk; (II) allocation concealment: none of the 10 studies mentioned whether allocation concealment was adopted, which indicated an unclear risk; (III) the blind method for the subjects: 7 of the 10 studies mentioned that the subjects signed informed consent forms, but it was unclear whether the blind method had been applied to the subjects, which indicated an unclear risk; (IV) the blind method for the assessors: none of the 10 references mentioned whether the blind method had been applied to the assessors, which indicated an unclear risk; (V) the complete data: all of the 10 studies had complete results data, which indicated a low risk; (VI) selective reporting: all of the 10 studies were free of selective reporting, which indicated a low risk; and (VII) other bias risks: 6 of the 10 studies had inconsistent populations between groups, which indicated a high risk, and it was not clear whether there were other bias risks in the remaining 4 studies, which indicated an unclear risk. Figures 2 and 3 showed the bias risk assessment results.

Surgery Time

In this study, 7 studies analyzed surgery time. The heterogeneity test results (I²=99%, P<0.00001) indicated a certain degree of heterogeneity; thus, the REM was used. The combined effect size was MD=−5.59; 95% CI, −15.26 to 4.09; Z=1.13; P=0.26, which suggested that the surgery time of holmium laser lithotripsy was shorter versus control group (Figure 4).

Length of Hospital Stay

There were 3 studies analyzing the length of hospital stay. The heterogeneity test results (I²=0%, P=0.42) indicated no heterogeneity; thus, the FEM was used. The combined effect size was MD =−0.26; 95% CI, −0.54 to 0.02; Z=1.85; P=0.06, which suggested that the hospital stay of patients treated by holmium laser lithotripsy was shorter versus the control group (Figure 5).

The Removal Rate

Four studies in the meta-analysis analyzed the calculus removal rate. The heterogeneity test results (I²=58%, P=0.07) indicated a certain degree of heterogeneity; thus, the REM was used. The combined effect size was MD =4.36; 95% CI, 0.95–20.07; Z=1.89; P=0.06, which suggested that the experimental group had a higher calculus removal rate (Figure 6).

Urinary Tract Infection

In this study, 5 studies analyzed postoperative urinary tract infection. The heterogeneity test results (I²=13%, P=0.33) indicated no heterogeneity; thus, the FEM was used. The combined effect size was MD =0.69; 95% CI, 0.49–0.98; Z=2.05; P=0.04, which suggested that the incidence of urinary tract infection in the experimental group was lower (Figure 7).

Mucosal Damage

In this study, 5 studies analyzed patients' postoperative mucosal damage. The heterogeneity test results (I²=0%, P=0.5) indicated no heterogeneity; thus, the FEM was used. The combined effect size was MD =0.67; 95% CI, 0.31–1.41; Z=1.06; P=0.29, which suggested that the mucosal damage of patients in experimental group was less severe (Figure 8).

Vesical Perforation

In this study, 6 studies analyzed the postoperative vesical perforation of patients. The heterogeneity test results (I²=0%, P=0.74) indicated no heterogeneity; thus, the FEM was used. The combined effect size was MD =0.9; 95% CI, 0.49–1.65; Z=0.34; P=0.74, which suggested that the vesical perforation in the experimental group was less severe (Figure 9).

Residual Calculi

In this study, 6 studies analyzed the postoperative residual calculi rate. The heterogeneity test results (I²=0%, P=0.84) indicated no heterogeneity; thus, the FEM was used. The combined effect size was MD =0.32; 95% CI, 0.18–0.58; Z=3.8; P=0.0001, which suggested that the residual calculi rate in the experimental group was lower (Figure 10).

Hematuria

In this study, 3 studies analyzed the postoperative hematuria of patients. The heterogeneity test results (I²=0%, P=0.13) indicated no heterogeneity; thus, the FEM was used. The combined effect size was MD =0.62; 95% CI, 0.33–1.15; Z=1.53; P=0.13, which suggested that the incidence of hematuria was lower in the experimental group (Figure 11).

Abdominal Pain

In this study, 3 studies analyzed the postoperative abdominal pain of patients. The heterogeneity test results (I²=0%, P=0.68) indicated no heterogeneity; thus, the FEM was used for the analysis. The combined effect size was (MD =0.78; 95% CI, 0.6–1.02; Z=1.84; P=0.07), which suggested that the incidence of hematuria was lower in the experimental group (Figure 12).

Publication Bias

As shown in Figure 13, hospital stay, calculus removal rate, postoperative urinary tract infection, mucosal damage, bladder perforation, residual calculi rate, hematuria, and abdominal pain were basically distributed within the CI, and article bias was low; however, the scattered points of surgery time were outside the CI, suggesting that the included articles had a certain publication bias.

Transl Androl Urol. 2021;10(8):3465-3475. © 2021 AME Publishing Company