Tranexamic acid for percutaneous nephrolithotomy

Percutaneous nephrolithotomy (PCNL) is the gold standard for the treatment of large kidney stones but comes with an increased risk of bleeding compared to other treatments, such as ureteroscopy and shock wave lithotripsy. Tranexamic acid (TXA) is an antifibrinolytic agent that has been used to reduce bleeding complications in other settings.

To assess the effects of TXA in individuals with kidney stones undergoing PCNL.

We performed a comprehensive literature search of the Cochrane Library, PubMed (including MEDLINE), Embase, Scopus, Global Index Medicus, trials registries, other sources of the grey literature, and conference proceedings. We applied no restrictions on the language of publication nor publication status. The latest search date was 11 May 2023.

We included randomized controlled trials (RCTs) that compared treatment with PCNL with administration of TXA to placebo (or no TXA) for patients ≥ 18 years old.

Two review authors independently classified studies and abstracted data. Primary outcomes were: blood transfusion, stone-free rate (SFR), and thromboembolic events (TEEs). Secondary outcomes were: adverse events (AEs), secondary interventions, major surgical complications, minor surgical complications, unplanned hospitalizations or readmissions, and hospital length of stay (LOS). We performed statistical analyzes using a random-effects model. We rated the certainty of evidence (CoE) according to the GRADE approach using a minimally contextualized approach with predefined thresholds for minimally clinically important differences (MCIDs).

We analyzed 10 RCTs assessing the effect of systemic TXA in PCNL versus placebo (or no TXA) with 1883 randomized participants. Eight studies were published as full text. One was published in abstract proceedings, but it was separated into two separate studies for the purpose of our analyzes. Average stone surface area ranged 3.45 to 6.62 cm². We also found a single RCT published in full text assessing the effects of topical TXA in PCNL versus placebo (or no TXA) with 400 randomized participants, the results of which are further described in the review. Here we focus only on the results of TXA used systemically. Blood transfusion - Based on a representative baseline risk of 5.7% for blood transfusions taken from a large presentative observational studies, systemic TXA may reduce blood transfusions (risk ratio (RR) 0.45, 95% confidence interval (CI) 0.27 to 0.76; I² = 28%; 9 studies, 1353 participants; low CoE). We assumed an MCID of ≥ 2%. Based on 57 participants per 1000 with placebo (or no TXA) being transfused, this corresponds to 31 fewer (from 42 fewer to 14 fewer) participants being transfused per 1000. Stone-free rate - Based on a representative baseline risk of 75.7% for SFR, systemic TXA may increase SFRs (RR 1.11, 95% CI 0.98 to 1.27; I² = 62%; 4 studies, 603 participants; low CoE). We assumed an MCID of ≥ 5%. Based on 757 participants per 1000 being stone free with placebo (or no TXA), this corresponds to 83 more (from 15 fewer to 204 more) stone-free participants per 1000. Thromboembolic events - There is probably no difference in TEEs (risk difference (RD) 0.00, 95% CI -0.01 to 0.01; I² = 0%; 6 studies, 841 participants; moderate CoE). We assumed an MCID of ≥ 2%. Since there were no thromboembolic events in intervention and/or control groups in 5 out of6 studies, we opted to assess a risk difference with systemic TXA for this outcome. Adverse events - Systemic TXA may increase AEs (RR 5.22, 95% CI 0.52 to 52.72; I² = 75%; 4 studies, 602 participants; low CoE). We assumed an MCID of ≥ 5%. Based on 23 participants per 1000 with placebo (or no TXA) having an adverse event, this corresponds to 98 more (from 11 fewer to 1000 more) participants with adverse events per 1000. Secondary interventions - Systemic TXA may have little to no effect on secondary interventions (RR 1.15, 95% CI 0.84 to 1.57; I² = 0%; 2 studies, 319 participants; low CoE). We assumed an MCID of ≥ 5%. Based on 278 participants per 1000 with placebo (or no TXA) having a secondary intervention, this corresponds to 42 more (from 44 fewer to 158 more) participants with secondary interventions per 1000. Major surgical complications - Based on a representative baseline risk for major surgical complications of 4.1%, systemic TXA may reduce major surgical complications (RR 0.36, 95% CI 0.21 to 0.62; I² = 0%; 5 studies, 733 participants; moderate CoE). We assumed an MCID of ≥ 2%. Based on 41 participants per 1000 with placebo (or no TXA) having a major surgical complication, this corresponds to 26 fewer (from 32 fewer to 16 fewer) participants with major surgical complications per 1000. Minor surgical complications - Systemic TXA may reduce minor surgical complications (RR 0.71, 95% CI 0.45 to 1.10; I² = 76%; 5 studies, 733 participants; low CoE). We assumed an MCID of ≥ 5%. Based on 396 participants per 1000 with placebo (or no TXA) having a minor surgical complication, this corresponds to 115 fewer (from 218 fewer to 40 more) participants with minor surgical complications per 1000. Unplanned hospitalizations or readmissions - We are very uncertain how unplanned hospitalizations or readmissions are affected (RR 1.55, 95% CI 0.45 to 5.31; I² = not applicable; 1 study, 189 participants; very low CoE). We assumed an MCID of ≥ 2%. Hospital length of stay - Systemic TXA may reduce hospital LOS (mean difference 0.52 days lower, 95% CI 0.93 lower to 0.11 lower; I² = 98%; 7 studies, 1151 participants; low CoE). We assumed an MCID of ≥ 0.5 days.

Based on 10 RCTs with substantial methodological limitations that lowered all CoE of effect, we found that systemic TXA in PCNL may reduce blood transfusions, major and minor surgical complications, and hospital LOS, as well as improve SFRs; however, it may increase AEs. We are uncertain about the effects of systemic TXA on other outcomes. Findings of this review should assist urologists and their patients in making informed decisions about the use of TXA in the setting of PCNL.