Acute kidney injury after robot‐assisted laparoscopic prostatectomy: A meta‐analysis

We investigated the rates of acute kidney injury (AKI) post robot‐assisted laparoscopic prostatectomy (RALP).

Improving Global Outcomes (KDIGO), Acute Kidney Injury Network (AKIN), and the Risk, Injury, Failure, Loss of kidney function, and End stage kidney disease (RIFLE). 6cause of the increased intraabdominal pressure from the pneumoperitoneum that results from carbon dioxide insufflation during RALP, changes in the cardiovascular and respiratory systems occur.These changes in haemodynamics may lead to AKI. 7 It is well established that AKI increases cardiovascular morbidity and mortality. 6Moreover, AKI can accelerate CKD progression, which is an independent risk factor for cardiovascular events. 8It has been found that the mortality rates are 4.8% after stage 1 AKI, 9.1% after stage 2 AKI, and 14.9% after stage 3 AKI.Postoperative AKI also leads to extended hospital stays and increased costs. 9e rates of AKI post-RALP are unknown.Previous metaanalyses comparing RALP with laparoscopic radical prostatectomy or open radical prostatectomy have not assessed AKI as a complication of RALP.The meta-analysis aimed to evaluate the rates of AKI post-RALP.

| MATERIALS AND METHODS
We conducted a comprehensive literature search on four electronic databases from inception to August 2023: PubMed, Embase, Google Scholar, and Scopus.The search keywords included a combination of 'Acute Kidney Injury', 'AKI', 'Robot-Assisted Laparoscopic Prostatectomy', 'Robotic Prostatectomy', 'Robotic-Assisted', 'Radical Prostatectomy', 'Prostatectomy'.A manual search of the bibliographic sections of the retrieved studies was performed for additional publications.The literature search strategy is presented in the supplementary data.
Two reviewers (D.C. and R.K.) independently examined the articles.Screening was done by reviewing titles and abstracts.After screening, the full texts of the articles were reviewed to determine if they should be included in our study.Inclusion and exclusion criteria were used to decide whether to include a study.Discussing and consulting a third author (A.P.) helped resolve reviewer disputes.Continuous variables were expressed as mean � standard deviation and categorical variables as percentages.A continuity correction of 0.001 was recorded when the number of events was zero in a study.

The inclusion criteria
The studies included in this analysis are considered to be a random sample drawn from a broader pool of potential studies; therefore, we adopted a random effects model.To calculate the pooled rates, mean estimates and the corresponding 95% confidence intervals (CIs), we applied the inverse variance random-effects DerSimonian-Laird method. 11,12Forest plots were drawn to visualize the results of all the statistical analyses.
Heterogeneity was estimated using two methods.First, we applied the Cochran Q statistic, which is a test of the null hypothesis that all the included studies share the same effect size.If true, then the expected value of Q would be equal to the degrees of freedom (number of studies minus 1).When the value of Q exceeds the degrees of freedom, we reject the null hypothesis.The alpha level for this test is conventionally set at 0.10 rather than 0.05 because the test has limited statistical power.When heterogeneity was identified by the Q statistic, we used the I 2 statistic.The I 2 statistic quantifies the percentage amount of variance in effect sizes that is not attributable to sampling error alone.For heterogeneity measured by I 2 statistics, values < 30%, 31%-60%, 61%-75%, and >75% were suggestive of low, moderate, substantial, and considerable heterogeneity. 13Statistical analysis was performed using Comprehensive Meta-Analysis software version 4 (Biostat).Publication bias was ascertained by visual inspection of the funnel plots and using Egger's test.
A p-value of <0.05 was considered significant and denotes publication bias.If publication bias was present, we used Duval and Tweedie's 'Trim and Fill' method to determine the impact of the bias.
The quality of the studies was assessed using the Newcastle-Ottawa scale (NOS).NOS evaluates studies based on three domains: a selection domain to determine how well the study samples were selected, a comparability domain to evaluate the controls in the study and an outcome domain to assess the adequacy of follow up assessment.This scale scores studies on a scale of 0-9, studies with a score ≥6 are of high quality, 3-5 are of medium quality and <3 are of low quality. 14,15Two authors (D.C. and R.K.) performed the quality assessment.
Ethics approval was not required for our meta-analysis because the data was already accessible to the public.

| RESULTS
A total of 284 citations were retrieved.After excluding duplicates, case reports, conference abstracts and review articles 162 were identified and excluded.Following the screening, 122 articles were selected for full-text review.Of these, the articles not meeting the inclusion criteria (n = 112) were excluded.Finally, 10 studies that fulfilled the prespecified criteria were selected for the meta-analysis.
The study selection flow chart is presented in Figure 1.
Nine of the studies included in the meta-analysis were retrospective observational studies [16][17][18][19][20][21][22][23][24] and one was a prospective study. 25Ning et al. examined the clinical outcomes of 136,790 patients with prostate cancer from the National Inpatient Sample (NIS) database undergoing either RALP or retroperitoneal radical prostatectomy (RRP).The cohort that underwent RALP was included in our meta-analysis. 23Five studies had a control group. 17,19,21,22,25The effects of nicardipine infusion on AKI were studied by Cho et al. 25

Ergin et al. and Joo et al. reported the changes in renal function after
RALP versus retropubic prostatectomy. 17,19Naito et al. compared AKI rates post RALP versus open radical prostatectomy. 22The effect of intra-operative mannitol infusion on AKI was reported by Kong et al. 21For our study, data were extracted only from the groups who did not undergo nicardipine or mannitol infusions.
Ahn et al. used Cockcroft and Gault's formula to calculate the estimated glomerular filtration rate. 16Modification of Diet in Renal Disease Study (MDRD) equation was used in 4 studies. 17,19,24,25ronic Kidney Disease Epidemiology Collaboration (CKD-EPI) was used in 2 studies. 20,21total of 7 studies used Kidney Disease Improving Global Outcomes (KDIGO) criteria to diagnose AKI.[17][18][19][20][21][22]24 In the study by Cho et al.AKI was defined as eGFR <60 mL/min/1.73 m 2 and hey also recorded the highest serum Cr level during the postoperative stay in hospital.25 The International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM) diagnosis codes were used to retrieve data from the NIS database by Ning et al. 23 This relied on the reported rates of AKI and did not have a specific AKI diagnosis definition criterion.

F I G U R E 1
Study selection process according to the preferred reporting items for systematic reviews and meta-analysis statement.
T A B L E 1 Summary of the included studies.The total number of included patients was 60,937.The mean pooled age was 65.1 years (95% CI: 62.8-67.4;I 2 : 99.4).The number of patients with reported comorbidities was 12.8% for type 2 diabetes, 35.8% for hypertension, 10.2% for coronary artery disease, and 7.4% for chronic kidney disease.Among the preoperative use of anti-hypertensive medications that was reported, 32.7% were on angiotensin converting enzyme inhibitors and angiotensin II receptor blockers.Beta-blockers were used in 5.9% and diuretics in 10.5%.The study characteristics are detailed in Table 1 and the details on intraoperative and post-operative data are summarised in Table 2.

Study Year Country Study design
The Newcastle-Ottawa scale was used to score the included studies.Nine studies were scored as high quality and one study was of medium quality.and 3.
Sensitivity analysis was performed to determine whether any single study had a dominant effect on the pooled AKI rate.One study was excluded at a time and the effect size was computed to note its effect on the main summary estimate.We determined that no single study had a dominant effect on the pooled AKI rate.Ning et al. reported the rates of AKI in 58,703 patients, although this cohort was large compared to other studies, 23 no significant effect was noted while performing sensitivity analysis.Q statistics and I 2 statistics were used to assess.The Q-statistic provides a test of the null hypothesis that all studies in the analysis share a common effect size.If all studies shared the same true effect size, the expected value of Q would be equal to the degrees of freedom (the number of studies minus 1).The I-squared statistic indicates that the variance in observed effects reflects variance in true effects rather than sampling error.The Q statistic detected the presence of heterogeneity in all our analyses.Using I 2 statistics, we determined that the presence of heterogeneity was >75% and so it was considerable.
Analysis of Publication Bias by visual inspection showed that there may be a publication bias due to the presence of asymmetry.

| DISCUSSION
In our meta-analysis, we report a pooled post-RALP AKI rate of 7.2%.
The included studies had a high variability in reporting the rates of AKI.Since various definitions exist to define AKI, we considered AKI was valid if the authors in the original study had reported it regardless of the specific criteria used to define it.
The occurrence of AKI post-RALP is controversial and is pri- The use of lung protective ventilatory strategies, positive endexpiratory pressure and optimal fluid management mitigates some of the effects of the Trendelenburg position.However, it should be noted that fluid restriction is sometimes employed due to concerns for laryngeal oedema, which could increase the risk for developing AKI. 27,28One study reported that the likelihood of AKI increases with prolonged pneumoperitoneum, and it can lead to a decline in renal function that persists up to 1 year. 24In contrast, another study reported that the renal parameters returned to baseline in 24 h post RALP and follow up of these patients showed no residual chronic kidney disease (CKD) 30 days after surgery. 16Nonetheless, it has been shown that this process may not be robust in all populations, especially in the elderly with a decreased renal reserve, thereby making them more susceptible to developing AKI. 28I could also result from blood loss during surgery, but RALP has been shown to have lower blood loss and transfusion requirements compared to retroperitoneal and laparoscopic radical prostatectomy. 2 A lower pressure pneumoperitoneum of ≤12 mmHg compared to a standard pressure pneumoperitoneum of >12 mmHg did not demonstrate increased estimated blood loss or other advantages, but the study did not assess the incidence of AKI. 7 CKD is a well-known risk factor for perioperative-AKI after cardiac and bariatric surgeries.29   reported that the risk of AKI was higher with retropubic radical prostatectomy (RRP) than with RALP after propensity matching with rates of 10.4% versus 5.5%.19 But Ergin et al. reported that the rates of AKI were higher in RALP than in RRP.However, there was significantly more blood loss post RRP. 17 To the best of our knowledge, this is the first meta-analysis assessing the rates of AKI after RALP. Tis meta-analysis includes data on 60,937 patients, which is the strength of the study.Also, the studies included had different methodologies and studied the occurrence of AKI in different populations.
Our study has a few limitations.First, most of the studies, but not all of them, used KDIGO criteria to diagnose AKI.Second, the metaanalysis included retrospective studies.Retrospective studies rely on the completeness of information recorded in medical records and could have missing or incomplete data.Our study inherits the bias and has the limitations of the included studies.Finally, there was considerable heterogeneity, so caution should be taken when interpreting the results of the meta-analysis and when extrapolating the findings to other populations.Nevertheless, this study adds valuable data to the current literature on this topic.
Clinicians should be aware of the possibility of AKI and institute better perioperative management in patients undergoing RALP.
Further studies are required to determine the long-term consequences of post-RALP AKI and to identify high-risk groups.
are: (1) Adults aged 18 years and older; (2) Availability of information on BUN, creatinine, or eGFR changes post-RALP and adequate follow-up postoperatively; and (3) Studies published in English.The exclusion criteria are: (1) Patients with a history of AKI prior to undergoing RALP; (2) Studies with a lack of sufficient data on renal parameters post-RALP; (3) Abstracts, case reports, and reviews; and (4) less than 10 patients in the study cohort.The Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines were used to identify the final articles. 10The included studies were verified not to contain any overlapping cohorts.The study protocol was registered in PROS-PERO (CRD42023452015).The primary outcome was the pooled postoperative rate of AKI after RALP.The mean operative time, anaesthesia time, estimated blood loss, intra-operative administration of fluids, and the length of stay were also assessed.Two authors (D.C. and R.K.) verified the extracted data after all the authors performed data extraction onto a standardized form.Country of origin, publication year, number of patients, demographic information, comorbidities, operative time, anaesthesia times, intraoperative fluids administered, blood loss during surgery, definition of AKI, length of stay, and study outcomes were extracted.Only the data pertaining to RALP were extracted if a study reported renal parameters after different methods of prostatectomy.

T A B L E 3 10 -
Figure 5.
marily thought to be related to the effects of pneumoperitoneum and the Trendelenburg position.An insufflation pressure of around 12-15 mmhg can lead to inferior vena cava compression, aortic compression, decreased splanchnic blood flow, and renal blood flow.Pneumoperitoneum has also been shown to increase systemic venous resistance (SVR) by almost 20%, which may cause a decrease in cardiac output.This decrease becomes more pronounced with very elevated pressures of >15-20 mm Hg.26A steep Trendelenburg position that is often used in urologic robotic surgery results in haemodynamic changes such as decreases in mean arterial pressures and cardiac output.

F I G U R E 2
Forest plot of AKI rate post-RALP.AKI, Acute kidney injury; CI, Confidence interval; RALP Robot assisted laparoscopic prostatectomy.F I G U R E 3 (A) Forest plot of mean anaesthesia time (minutes) during RALP, (B) Forest plot of mean operation time (minutes) during RALP, (C) Forest plot of mean intraoperative fluids administered (ml) during RALP, (D) Forest plot of mean estimated blood loss (ml) during RALP, (E) Forest plot of mean length of stay (days) post-RALP.AKI, Acute kidney injury; CI, Confidence interval; RALP Robot assisted laparoscopic prostatectomy.

Table 3
A large study byNing et al. using 23LP remains a valuable minimally invasive surgical option due to its many benefits and since other methods of radical prostatectomy may not, in fact, confer lower rates of AKI.Patients who underwent RALP had a decreased risk of AKI and other postoperative complications compared to open prostatectomies (ORP).23Jooet al.