Two independent investigators searched PubMed, Web of Science, Embase, Cochrane Library, China National Knowledge Infrastructure (CNKI), Wanfang Data, and Viper Information Products databases. The search was conducted using a combination of subject terms and free words. English search terms: “Ureteroscopy/Lithotripsy/Ureteroscopy stone surgery/Ureteroscopic stone removal/Flexible ureteroscopy/Ureterolithotripsy/Ureteroscopic surgery/Ureteroscopic treatment,” “urosepsis/Urinary sepsis/ Septic shock/Urinary tract infection sepsis/Sepsis,” and “Risk factors/Influencing factors/Relevant factors/Associate factors/Predictive factors.” Additionally, references of the retrieved articles were manually reviewed to identify further relevant studies. The search covered the period from the inception of each database to January 31, 2025.

Inclusion criteria were as follows: •

Subjects: Patients undergoing URSL

Two researchers skilled in meta-analysis methods independently removed duplicates using EndNote 20 and then screened the titles and abstracts of the remaining documents to eliminate those that did not meet the eligibility criteria. They reviewed the full text to determine the final inclusion of the selected reports. The screening results were cross-checked, and discrepancies were resolved by discussion; a third researcher adjudicated unresolved disagreements. Information about the relevant literature was extracted, including: first author, year of publication, country, type of study, sample size, number of patients with urosepsis, influencing factors, OR, and 95% confidence (CI).

The quality of bias in case-control and cohort studies was assessed using the Newcastle-Ottawa Scale (NOS) (9). This scale evaluates literature quality across three dimensions, comprising a total of eight entries: four entries for participant selection, one for group comparability, and three for outcome assessment. While each standard entry is scored on a 1-point scale, the comparability item allows for a maximum of 2 points, resulting in a total possible score of 9 points across all domains. Studies were categorized as low (0–4 points), medium (5–7 points), and high quality (>7 points). Two researchers independently evaluated the studies, and any differences in scoring were resolved through discussion, with the help of a third researcher.

Stata 18.0 was used for meta-analysis, subgroup analysis, and sensitivity analysis, while Egger's test was performed using Stata 15.1. All data were obtained directly from the original studies without any processing or transformation. The statistical results were presented as OR with 95% CI. Heterogeneity was evaluated using Cochran's Q test and I² statistics. A fixed-effects model was applied when I² < 50% and P > 0.10; otherwise, a random-effects model was used when I² ≥ 50% and P ≤ 0.10. Subgroup analyses and sensitivity analyses were conducted to explore potential sources of heterogeneity. Funnel plots and Egger's test were used to assess publication bias when there were at least 10 studies, and the trim-and-fill method was applied when bias was detected. Statistical significance was set at P < 0.05.

The initial search identified 1,152 potentially relevant articles. After a stepwise screening process, 26 studies (10–27) met the inclusion criteria, consisting of 16 articles published in English and 10 published in Chinese. The combined study population included 12,394 patients, of whom 861 cases were diagnosed with urosepsis. The literature screening process and results are illustrated in Figure 1.

The analysis identified 36 risk factors across all included studies, with 13 factors reported in at least two studies. Based on the NOS quality assessment, 13 studies were rated medium quality and 13 high quality; none of the articles were of low quality. The characteristics of the included studies and the results of the quality assessment are shown in Table 1.

Meta-analysis was conducted for factors reported in at least two studies, including two demographic factors (gender, age), two stone-related factors (stone size, stone number), six infection-related factors [history of UTI, positive urine culture, positive urine nitrite, leukocyte count, C-reactive protein (CRP), procalcitonin], two surgical factors (operation time, stent placement), and one comorbidity-related factor (diabetes mellitus).

Stone size, stone number, history of UTI, positive urine culture, positive urine nitrite, CRP levels, operation time, stent placement, and diabetes mellitus were significantly associated with urosepsis following URSL (P < 0.05). However, no significant associations were found for gender, age, leukocyte count, or procalcitonin levels (P > 0.05), as shown in Table 2.

A total of 22 factors were reported in single studies, including: stone-related factors (stone location, residual stones, stone composition, and stone obstruction); procedure-related factors (stent retention time, irrigation pressure, intraoperative fluid volume, and antibiotic use); laboratory markers (albumin levels, albumin variability, albumin-globulin ratio, urinary leukocyte count, neutrophil-lymphocyte ratio, blood urea nitrogen, platelet count, heparin-binding protein level, and serum creatinine); underlying diseases (neurological disorders and history of urological surgery); and other factors (qSOFA/SOFA score, fever, and hydronephrosis). The clinical significance of these factors requires cautious interpretation due to limited evidence from single studies.

For factors with I² > 50%, subgroup analyses were conducted by sample size, study design, and number of urosepsis cases to identify potential sources of heterogeneity. When subgroup analysis was not feasible or significant heterogeneity persisted after subgroup analysis, sensitivity analysis was performed by sequentially excluding individual studies to assess changes in the combined OR values and P values, helping to identify possible sources of heterogeneity.

The heterogeneity test for gender showed significant variation across studies (I² = 81%, P = 0.001). As subgroup analysis was not feasible, we performed sensitivity analysis by sequentially excluding individual studies (Figure 2). The study by Xiaobo et al. (12) was identified as the primary source of heterogeneity. After excluding this study, heterogeneity was significantly reduced (I² = 0%, P = 0.911), and a fixed-effects model was applied. The analysis revealed that gender was significantly with higher risks of urosepsis following URSL (OR = 2.90, 95% CI: 1.89, 4.46; P < 0.001).

The heterogeneity test for stone size showed significant variation across studies (I² = 92.9%, P = 0.001). Subgroup analyses by sample size, stone diameter, and number of urosepsis cases (Table 3 and Figure 3) revealed significant between-group differences by sample size (P = 0.001) and stone diameter (P = 0.032), suggesting these factors may influence urosepsis risk after URSL. However, substantial heterogeneity persisted within these subgroups, indicating that other sources of variation remain unaccounted for.

For history of UTI, there was heterogeneity across studies (I² = 80.7%, P = 0.001). Subgroup analysis by sample size and study design (Table 4 and Figure 4) showed statistically significant differences by sample size subgroups (P = 0.032). Heterogeneity was significantly reduced in both subgroups (≤500 group: I² = 0%, P = 0.406; >500 group: I² = 30.8%, P = 0.229), suggesting that sample size may influence the association between UTI history and risk of urosepsis after URSL. Fixed-effects model analysis showed a statistically significant pooled effect (OR = 5.96, 95% CI: 4.12, 8.60; P < 0.001).

For operative time, there was significant variation across studies (I² = 88.9%, P < 0.001). Subgroup analyses by sample size, study design, number of urosepsis cases, and publication year did not significantly reduce heterogeneity (Table 5 and Figure 5). Sensitivity analysis by sequentially excluding individual studies identified five studies (11, 12, 16, 17, 28) as the primary sources of heterogeneity. After excluding these studies, heterogeneity was significantly reduced (I² = 43.8%, P = 0.113). Fixed-effects meta-analysis demonstrated a statistically significant pooled effect (OR = 1.09, 95% CI: 1.07, 1.11; P < 0.001).

For factors reported in more than 10 studies, publication bias was assessed using funnel plots (Figure 6 and Figure 7) and Egger's test (Table 6). The findings revealed potential publication bias for operation time (P = 0.001) and positive urine culture (P = 0.013). The trim-and-fill method was applied using Stata 15.0 to correct for potential bias.

For surgical time, the adjusted effect size (OR = 1.15; 95% CI: 0.96, 1.37; P = 0.133), differing substantially from the original estimate (OR = 0.48, 95% CI: 0.26, 0.57); P < 0.001), suggesting significant publication bias, indicating that the original results may have been affected by unpublished studies. In contrast, the adjusted effect size for positive urine culture (OR = 3.77, 95% CI: 3.07, 4.63; P < 0.001) was relatively consistent with the original findings (OR = 1.60 95% CI: 1.37, 1.83; P < 0.001), suggesting the initial results were robust despite some publication bias.

The findings of this meta-analysis demonstrated that diabetes mellitus was significantly associated with urosepsis post-URSL, consistent with the findings of Sun et al. (29), which revealed an association between diabetes mellitus and the risk of infection. Hyperglycemia in diabetes mellitus not only provides an energy source for bacterial growth but also significantly dysregulates both innate and adaptive immune responses, increasing susceptibility to infections. Innate immune dysfunction includes impaired neutrophil chemotaxis, phagocytosis, and oxidative burst, reduced complement activation, and NK cell cytotoxicity. Adaptive immunity is suppressed by defective antigen-presenting cell (APC) maturation, diminished cytokine production, CD4+ T-cell dysfunction, and disrupted B-cell antibody production (30, 31). These immune and metabolic alterations account for the increased infection risk in diabetic patients following URSL.

Although gender was not statistically significant in this meta-analysis, Dan et al. (32) have reported that gender significantly impacts the risk of infection following URSL. Potential reasons for women's increased susceptibility to urosepsis after URSL may include genetic factors and effects of sex hormones (29). Genetic differences can affect immune system function and infection resistance by altering how immune-related genes work. Female sex hormones like estrogen and progesterone may also influence inflammation and immune responses by changing how immune cells function and produce cytokines, which could affect the body's ability to fight infections. Assessment of preoperative infection risk is essential in female patients and patients with diabetes undergoing URSL. Appropriate preventive measures, such as glycemic control, should be implemented to reduce the risk of postoperative urosepsis.

The meta-analysis found that stone size and stone number were significantly associated with the risk of urosepsis following URSL. Although flexible ureteroscopy has been demonstrated to be effective for renal calculi >2 cm (33), Jeong et al. (34) found that stone diameter ≥1.5 cm is associated with increased infection risk. The duration of lithotripsy is significantly prolonged in cases involving larger or multiple urinary stones. Continuous laser energy and mechanical forces break down the stone structure, releasing trapped endotoxins and bacteria (35). During the procedure, the necessary high-pressure irrigation increases the permeability of small blood vessels and lymph channels of the kidney, creating pathways for bacteria to enter the bloodstream. Additionally, larger stones can cause urinary tract obstruction (36), impairing urine drainage and promoting bacterial growth. These physiological changes allow bacteria and their toxic products to spread throughout the body, significantly increasing the risk of serious infection. For patients with large or multiple stones, preoperative infection risk assessment is crucial, and lithotripsy time should be minimized to reduce postoperative infections.

This study found that a history of UTI, positive urine culture, positive urinary nitrite, and elevated CRP levels were significantly associated with urosepsis following URSL. Persistent fever typically indicates the presence of pro-inflammatory factors and uncontrolled infection. Inflammation in the urinary tract induces mucosal edema, vasodilation, and compromised vascular barrier function, thereby disrupting the natural mucosal defense and increasing the risk of bacterial and toxin translocation into the circulation. Francesco et al. (37) demonstrated that prolonged URSL operation time is significantly associated with an elevated risk of postoperative UTI. Urine culture positivity is the gold standard for diagnosing UTI, providing direct evidence of viable bacteria in the urine. A meta-analysis by Naeem et al. (38) demonstrated that a positive urine culture is a risk factor for sepsis, with microbiological analysis identifying Escherichia coli as the predominant uropathogen (39). A positive nitrite result indicates the reduction of nitrate to nitrite by Gram-negative bacteria in urine. This biochemical conversion typically requires a high bacterial load, and such significant bacteriuria frequently triggers pronounced inflammatory responses. CRP serves as a sensitive systemic inflammation marker that can increase up to 1,000-fold at infection or inflammation sites (40). The non-significant association for procalcitonin in this study may be attributed to the limited number of included studies. Villanueva et al. (41) reported that leukocyte counts typically increase during bacterial UTIs. The inconsistent findings in this study may be because severe infections can show an initial decrease followed by an increase due to endotoxin-mediated leukocyte depletion (42). Notably, the associations for both procalcitonin (OR = 9.34, 95% CI: 0.1–878.49; P = 0.335) and positive urine nitrite (OR = 7.68, 95% CI: 1.03–52.27; P < 0.001) had markedly wide confidence intervals, reflecting population heterogeneity and methodological differences. Thus, we strongly recommend cautious interpretation of these biomarkers in clinical practice. Therefore, UTIs should be treated before lithotripsy. Antibiotic therapy should be guided by preoperative urine culture (43), and prophylactic antibiotics may also help reduce infection risk in patients with negative urine cultures.

We found that procedure duration and stent placement were significantly associated with urosepsis post-URSL. Ozgor et al. (44) reported that procedures lasting more than one hour doubled the risk of infectious complications, as prolonged exposure increases the likelihood of pathogenic bacteria and endotoxins entering the bloodstream. Preoperative double J stent placement improves stone clearance in patients undergoing URSL. However, a meta-analysis by Corrales et al. (45) demonstrated that stent retention exceeding 30 days increases sepsis rates. Prolonged stenting may induce chronic inflammation, causing local tissue damage and persistent release of inflammatory factors, thereby elevating sepsis risk. Therefore, although preoperative stent placement is recommended, its duration should be carefully monitored. Additionally, controlled procedure times are essential during surgery.

This study has several limitations: (1) The inclusion of only Chinese and English literature may introduce language bias; (2) Most included studies were of moderate quality, highlighting the need for additional high-quality research; (3) Significant heterogeneity was observed in some influencing factors—although subgroup analyses identified partial sources, other heterogeneity factors remain to be elucidated; (4) Some risk factors were reported in single studies, limiting the reliability of these findings due to insufficient data.