Comparison of safety and efficacy of different positions in percutaneous nephrolithotomy: a network meta-analysis

Abstract

Objectives:

Various new positions for percutaneous nephrolithotomy (PCNL) were proposed to reduce the limitations of the traditional position. This study was aimed to evaluate the efficacy and safety of the different PCNL positions.

Methods:

PubMed, Embase, Web of Science, and the Cochrane Library were searched for relevant randomized controlled trials (RCTs) up to 18 April 2023. The authors collected five common surgical positions used for PCNL: oblique supine position (OSP), supine position (SP), flank position (FP), split-leg oblique supine/flank position (SLP), and prone position (PP). Paired and network meta-analysis were conducted to compare relevant outcomes, including complications, operative time, stone-free rates, hospital stay, and hemoglobin loss among these different positions.

Results:

The study included 17 RCTs with a total of 1841 patients. The result demonstrated that SLP significantly outperformed in terms of decreasing operation time (FP vs SLP MD- MD-41.65; OSP vs SLP MD 28.97; PP vs SLP MD 34.94), hospital stay, and hemoglobin loss. Ranking probabilities showed SLP had highest stone-free rate. Prone position was more likely to occur complications than others. Based on SMAA model, the benefit-risk analysis suggested the SLP was the optimal position in PCNL.

Conclusions:

For PCNL, the split-leg, flank, supine, and OSPs are as secure as the prone position. Further RCTs are necessary to confirm the outstanding safety and efficacy of split-leg position. Besides, the position should be selected regard for the patient’s demands, the surgeon’s preference and learning curve.

Introduction

Highlights

• The classic prone position is no longer the optimal choice for percutaneous nephrolithotomy (PCNL), potentially due to the global rise in obesity rates.

• The split-leg oblique supine/flank position in PCNL surgical positioning has demonstrated a higher stone clearance rate and improved safety.

• The learning curve of surgeons significantly influences the selection of PCNL positioning.

Percutaneous nephrolithotomy (PCNL) was first made available as a kidney stone therapy in 1976¹. PCNL has become the typical approach to treating patients with kidney stones bigger than 2 cm in diameter, particularly those with staghorn stones^2,3. The choice of PCNL operative position is very important. Both the pursuit of adequate space for surgical manipulation and the reduction of anesthetic risks and respiratory circulation complications are pursued by all surgeons. Therefore, clinical practice has proposed a number of surgical positions for optimizing the efficiency and security of PCNL⁴.

The prone position was the initial position and remains one of the prevailing choices in contemporary practice⁵. Although prone position provided suitable operative space and low abdominal damage rate, there are some drawbacks. For patients with cardiac comorbidities, morbid obesity, skeletal deformities, the prone position is contraindicated. Additionally, Valdivia-Uria and colleagues described a lateral approach to perform PCNL in the supine position in 1998⁶. The supine position is gradually gaining prominence due to the advantages it offers in Endoscopic Combined Intrarenal Surgery (ECIRS) and under general anesthesia, challenging the conventional prone position⁷. In 2011, Lezrek’s team introduced the split-leg position⁸. Several other positions, including the oblique supine and flank positions, have been proposed to enhance surgical technique⁹.

Various positions in PCNL offer distinct advantages and limitations, which may vary depending on the characteristics of different patient populations. The rising global prevalence of obesity increased the incidence of circulatory and respiratory complications following anesthesia, further complicates the selection of the optimal PCNL position¹⁰. Despite numerous comparative studies investigating PCNL positions, a conclusive consensus remains elusive, emphasizing the need for further research in this area².

Based on the specific placement and angle of inclination, we divided all positions into five categories, comprising the supine position, oblique supine position, flank position, split-leg oblique supine/flank position, and prone position. Network meta-analysis (NMA) is suitable to synthesize and compare those positions at the same time. NMA allowed us to explore not only direct comparisons between these positions, but also indirectly to compare interventions that have not been directly pitted against each other in clinical trials. To evaluate the security and effectiveness of different PCNL positions and consider the learning curve for surgeons, we carried out a NMA and systematic review in this study.

Methods

Study design

We conducted NMA in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-analysis (PRISMA) standard and assessing the methodological quality of systematic reviews (AMSTAR) (Supplementary Table S1, Supplemental Digital Content 1, http://links.lww.com/JS9/C14 and Supplementary Table S2, Supplemental Digital Content 2, http://links.lww.com/JS9/C15)^11,12. The study has been registered in PROSPERO.

Search strategy

We searched for randomized controlled trials (RCTs) with no date restrictions before 18 April 2023 in PubMed, EMBASE, Cochrane, and Web of Science (WOS). Keywords included ‘percutaneous nephrolithotomy’, ‘PCNL’, ‘position’, and ‘randomized controlled trial’. The search strategy used in PubMed is ‘((((Nephrolithotomy, Percutaneous[MeSH Terms]) OR (PCNL)) OR (Percutaneous Nephrolithotomies)) OR (Percutaneous Nephrolithotomy)) AND (position) AND ((Randomized Controlled Trial[Publication Type]) OR (randomized controlled trial))’. More details on other databases were described in Supplementary Table S3 (Supplemental Digital Content 3, http://links.lww.com/JS9/C16). Despite a systematic evaluation of selected studies’ references, no additional publications were found.

Each article identified by the electronic search underwent relevance screening by two authors independently. Titles and abstracts of relevant publications were evaluated first, and then the entire texts of the selected articles were examined. We specifically searched for peer-reviewed studies, and did not consider papers in languages other than English, gray literature, or registry studies. A third senior author was brought in to settle any disagreements that arose throughout the systematic literature review.

Criteria for inclusion and exclusion

The following is a description of our detailed inclusion criteria based on the PICOS principles:

1. Participants: Patients diagnosed with nephrolithiasis and suitable for PCNL therapy.

2. Interventions: Only studies focused on PCNL positions will be considered.

3. Comparisons: Studies must include comparative analyses of different PCNL positions.

4. Outcomes: Publications should report on relevant outcomes, including operative time, stone-free rate, occurrence of complications, length of hospitalization, and hemoglobin loss.

5. Study Design: Only RCTs will be eligible for inclusion in this study.

We followed these criteria to ensure the selection of high-quality studies that meet our research objective. The exclusion criteria were those that did not meet the above requirements. Non-RCTs, previously published systematic reviews, letters, comments, and conference abstracts were not included in the study. Low-quality literature with no baseline metrics and no key endpoints will also be excluded.

Data collection

After carefully examining the original position descriptions, we categorized all the positions described by the authors into five main posture categories: supine position (SP), oblique supine position (OSP), flank position (FP), split-leg oblique supine/flank position (SLP), and prone position (PP). The schematic figure of these five positions is presented in Supplementary Figure S1 (Supplemental Digital Content 3, http://links.lww.com/JS9/C16).

Two trained investigators used standardized forms to extract the following information: year of publication, total number of participants, number of male and female participants, participants’ age, their BMI, stone size, and stone-clearing criteria for each group. The primary outcomes evaluated were the stone-free rate, hospital duration of stay, complications, operation time, and hemoglobin loss. In cases when sample mean and SD are not provided in the article, we used a freely available web tool (https://www.math.hkbu.edu.hk/tongt/papers/median2mean.html) to generate the missing information^13,14. The sample size, median, range, and interquartile range from the source article will serve as the basis for this estimate.

The data extraction and management process, which were based on predefined criteria involved two independent authors. To ensure consistency, inter-rater reliability checks were performed periodically during the process. Any discrepancies between the two independent authors during data extraction were resolved by a third senior author with expertise in the field through discussion and consensus. Any discrepancies identified were addressed promptly.

Quality assessment

Two reviewers independently assessed the risk of bias using an instrument developed by the Cochrane Collaboration. The applied tools evaluated and took into account a wide range of potential forms of bias, including those associated with selection, performance, detection, attrition, and reporting.

Data synthesis

This study utilized ADDIS V1.16.8 and R v4.0.5 (R Foundation for Statistical Computing), including ‘Gemtc’ and ‘rjags’ for data analysis. The outcomes of the paired-meta comparisons were displayed using forest plots. An I² value greater than 50% was considered indicative of substantial heterogeneity. Funnel plot was used to observe publication bias. In the NMA research, the node-split analysis was performed to determine whether or not the closed-loop network was consistent when compared directly and indirectly, which used a Bayesian random effects model. If the differences were not statistically significant (P>0.05), then a consistency model was used, while an inconsistency model was used due to high heterogeneity when there was a significant difference. A potential scaling reduction factor within 1.02 was considered acceptable. To determine the relative merits of different PCNL positions, a benefit-risk analysis was performed using ADDIS and stochastic multicriteria acceptability analysis (SMAA) models. The SMAA model is a robust framework designed to evaluate multiple criteria for decision-making, particularly useful in contexts where uncertainties and preference information are prevalent^15,16. We used Grading of Recommendations Assessment, Development and Evaluation (GRADE) to assess the certainty of evidence^17,18. Microsoft Excel was used to visualize the ranking probabilities. The mean difference (MD) was utilized for continuous outcomes, while the ratio of ratios (OR) was applied for categorical variables with 95% CI. The significance threshold for the two-sided test was less than 0.05 for the P-value.

Results

Search results

Initially, an electronic search yielded a total of 210 records, of which 113 were removed as duplicates, and 76 were rejected after screening of the title and abstract. Two nonrandomized control studies and one postoperative research, as well as another study that did not fulfill the requisite outcome indicator, were omitted from the meta-analysis after review of the remaining literature. We visually represented the literature screening process and record retention with a PRISMA flowchart (Fig. 1). Ultimately, we included 17 RCTs with a total of 1841 patients in our NMA^19–35.

Figure 1.

PRISMA flow diagram showing selection of studies.

Basic characteristics and risk of bias

One publication used a mix of fluoroscopy and ultrasound for guidance³¹, while five other papers reported using ultrasound guidance alone or in conjunction with fluoroscopy for PCNL operations^{19,22,23,26,30}. With the exception of the study conducted by Desoky E²², the research participants were adults with a BMI above the healthy range of 24. We outlined the primary features of the collected researches (Table 1).

Table 1.

Characteristics of included randomized controlled trials.

No	Author (Year)	Surgical method	Surgical position	Sample size (male/total)	Mean age (Year)	BMI	Stone size (mm)	Research type	Stone-free standard	Outcome
1	Ahmed et al. 202124	Percutaneous nephrolithotomy (PNL) Laser lithotripsy	Prone position vs split-leg modified lateral position	(38/61) vs (45/63)	46.49±13.47 vs 45.62±9.47	28.62±4.02 vs 27.09±2.77	28.34±10.02 vs 32.87±4.61	A prospective, randomized, unblind, double arm trial	Stone-free status was defined as no visualized stone or residual fragments <3 mm on postoperative abdominal NCCT	Stone-free rate (SFR), total operative time, track formation time, fluoroscopy time, auxiliary procedures, and complication rates, hemoglobin reduction
2	Al-Dessoukey et al. 201429	Percutaneous nephrolithotomy (PCNL) Laser lithotripsy	Prone position vs the split-leg oblique supine lithotomy position	(68/101) vs (68/102)	34.86±18.967 vs 37.21±14.526	27.24±3.56 vs 26.87±3.41	36.8 ±14.2 vs 39.3±12.6	A randomized comparative study	Stone size <4 mm	Stone-free rate, blood transfusion rate, and complication rates, mean hemoglobin loss, mean operative time, mean hospital stay, anesthesiological parameters
3	De Sio et al. 200835	Percutaneous nephrolithotomy Ultrasonic lithotripsy	Supine position vs prone position	(17/39) vs (16/36)	38 (25–72) vs 41 (28–69)a	28 (24–30) vs 26 (23–30)a	34 (25–51) vs 33 (27–45)a	A prospective randomized trial	Patients were considered stone-free when no stone >2 mm was visualized	Stone-free rate, mean blood loss, mean hospital stay, mean operative time. No blood transfusions were needed and no organ injuries were reported
4	Desoky et al. 202222	Ultra-mini-percutaneous nephrolithotomy (UMPCNL) Laser lithotripsy	Flank-free modified supine (FFMS) vs prone positions	(18/28) vs (19/27)	9.5±3.1 vs 10.3±3.4	22.2±2.16 vs 21.9±2.53	24.7±2.32 vs 25.1±2.12	A prospective randomized study	If the patient was stone-free or had residual fragments <4 mm	Operation time, stone-free rate, overall complication rate, postoperative pain, hospital stay
5	El-Shaer et al. 201926	Ultrasound PCNL Ballistic lithotripsy	Supine position vs Prone position	(91/129) vs (89/132)	38.8±11 vs 39.6±9.6	28.8±3.4 vs 27.9±3.2	32.2±9vs 31.2±8.9	A prospective randomized and controlled study	NR	Operative time, hospital stay, perioperative morbidities, stone-free rate, the mean nephrostomy time, the mean percentage decrease in hemoglobin concentration, complication
6	Falahatkar et al. 200834	Percutaneous nephrolithotripsy (PCNL) Ballistic lithotripsy	Complete supine position vs prone position	(23/40) vs (18/40)	45.35±11.43 vs 43.02±13.08	25.6±3.5 vs 26.3±4.2	40.6±15.4 vs 40.3±16.3	A randomized study	We defined stone 5 mm as stone-free rate	Stone-free rate, Mean hospital stay, Complications
7	Falahatkar et al. 201133	PCNL percutaneous nephrolithotripsy	Complete supine vs prone position	(15/18) vs (10/15)	49.9±12.4 vs 47.06±7.5	26.9±3.4 vs 24.8±3.07	31.2±14.5 vs 27.3±6.9	A randomized controlled trial	Stone-free rate was defined residual stones less than 5 mm	Operative time, stone-free rate, complications and hospital stay
8	Falahatkar et al. 201728	Percutaneous nephrolithotomy (PCNL) Ballistic lithotripsy	Complete supine vs semi supine	(8/22) vs (10/22)	52.59±11.77 vs 47.55±12.92	27.41±4.11 vs 27.07±5.18	NR	A randomized clinical trial controlled	Stone-free status was considered as residual stone less than 4 mm	Operative time, stone-free rate and complication
9	Giusti et al. 202025	Percutaneous nephrolithotripsy (PCNL) Ballistic lithotripsy	Modified supine position (‘Double-S’) vs prone position	(29/45) vs (25/45)	53.06±13.71 vs 51.133±15.23	25.555±3.724 vs 24.588±3.257	NR	A prospective randomized clinical trial	Stone-free status<3 mm	Complications, hemoglobin loss, Length of Hospital stay days (mean), Stone-free rate
10	Hosseini et al. 202221	Percutaneous nephrolithotomy (PCNL) in patients with a BMI above 30. Ballistic lithotripsy	Flank vs prone position	(16/31) vs (18/29)	47.45±7.16 vs 47.70±9.24	31.40±1.09 vs 31.00±0.91	3.20±0.69 vs 3.13±0.97	A randomized clinical trial	The remaining residual below 3 mm were considered as success rates	Arterial blood gases, Mean stone size, arterial blood gases, SFR (stone-free rate), Complications
11	Karami et al. 201332	Percutaneous nephrolithotomy (PCNL) Ballistic lithotripsy	Prone vs supine position	(34/50) vs (31/50)	44.4 ± 9.4 vs 41.5 ± 8.8	27.8 ± 4.3 vs 26.1 ± 4.1	28.2 ± 4.1 vs 28.3 ± 3.6	A randomized trial	Residual stones of less than or equal to 3 mm in diameter	Success rates, mean operation time, mean access duration, pyelocaliceal perforation
12	Perrella, et al. 202220	Percutaneous nephrolithotomy (PCNL) Ultrasonic lithotripsy	Supine position [Barts flank-free modified position] vs prone position	(23/56) vs (28/56)	50.3±12.4 vs 52.0±13.4	27.8±4.5 vs 28.6±4.9	60.1±25.8 vs 59.3±5.2	A noninferior randomized controlled trial	The absence of residual fragments (RFs) >4 mm	Success rates, final SFRs, operative time, complications
13	Radfar et al. 202123	Ultrasonic-guided PCNL Ballistic lithotripsy	Flank vs prone position	(61/100) vs (54/100)	42.34±6.62 vs 44.03±7.22	26.64±6.32 vs 25.11±5.18	27.12±2.98 vs 27.82±3.35	A randomized clinical trial	Success rate was defined as stone-free rate plus stone burden <4 mm	Success rate, operative time, access time, hemoglobin and creatinine changes, hospital stay, auxiliary procedure and surgical complications
14	Seleem et al. 202219	Ultra-mini-percutaneous nephrolithotomy (UMP) Laser lithotripsy	Modified flank free supine position vs prone position	(34/61) vs (32/61)	40.09±13.63 vs 39.67± 13.80	28.2±3.4 vs 27.5±3.1	15.1±3.5 vs 14.8±2.9	A Randomized Clinical Trial	The stone-free rate was considered if no fragments were left or small fragments <4 mm were present	operative time, fluoroscopy time, operative and postoperative complications, hospital stay, initial stone-free rate
15	Sofer et al. 201727	Fluoroscopic-guided PCNL Ultrasonic and ballistic lithotripsy	Oblique supine vs prone position	(9/27) vs (9/24)	59.5 vs 56.8b	30.9 vs 28.4b	30.9 vs 30.8	A randomized controlled trial (RCT)	NR	Operative time, anesthesia time, hospitalization, stone-free rates (SFR)
16	Wang et al. 201331	Fluoroscopy and ultrasound-guided PCNL procedures Ultrasonic lithotripsy	Modified supine positions vs prone positions	(28/60) vs (34/62)	44(30-69) vs 42(22-70)a	24(21-28) vs 25(20-28)a	NR	A prospective randomized study	A stone-free state was defined as no residual stones of diameter >4 mm.	Operation time, stone-free rate, loss of blood, hospital stay, second phase PCNL
17	Zhan et al. 201330	Minimally invasive percutaneous nephrolithotomy (MPCNL) with ultrasound Laser lithotripsy	Supine lithotomy vs prone position	(36/53) vs (38/56)	45±13.0 vs 44±15.0	24±3 vs 25±4	33±4 vs 34±5	A randomly controlled study	Patients were considered stone-free when no stone >2 mm was visualized in KUB	Stone-free rate, mean blood loss, number of access tracts, calyx puncture, mean hospital stay, complications, operative time

^aIQR, interquartile range.

^bNo specific SD was provided in the original article.

The quality of all included studies was evaluated using the Cochrane GRADE tool (Supplementary Figure S2, Supplemental Digital Content 3, http://links.lww.com/JS9/C16). Due to the complexity of blinding for surgical positions, the risk of performance bias was elevated. In two of the studies^30,33, the continuous variables in the results may be presented in an inappropriate way with too large SD indicated that data is non-normally distributed, indicating a high potential for bias. Other elements are supplemented in Supplementary Table S4 (Supplemental Digital Content 3, http://links.lww.com/JS9/C16).

Outcome for paired meta-analysis

Paired meta-analysis was conducted before the NMA. The outcomes in Supplementary Figure S3 (Supplemental Digital Content 3, http://links.lww.com/JS9/C16) indicated that the supine position was associated with a substantially shorter operative time than the prone position [SMD 1.40, 0.98–1.99] with high heterogeneity. The remaining results have no statistical significance.

NMA overview

Our NMA resulted in a network map depicting the relationships between the various PCNL positions (Fig. 2). Node-splitting modeling assessed direct-indirect comparison consistency. Summary tables were generated to assess consistency, and no inconsistencies were detected (Supplementary Table S5, Supplemental Digital Content 3, http://links.lww.com/JS9/C16). Hemoglobin Loss did not form a loop. We applied a consistency model for all metrics.

Figure 2.

Network diagram between different percutaneous nephrolithotomy positions.

Efficacy comparison

Seventeen RCTs include 898 patients in prone position, 243 in oblique supine position, 407 in supine position 131 in flank position and 162 in split-leg position were pooled together for Stone-free rates using NMA^19–35. Stone-free rates did not differ significantly between the different position groups (Fig. 3A). In the rank order, the outcomes demonstrated that the stone-free rate was highest for the split-leg position, followed by the supine position, the flank position, the prone position, and the oblique supine position had the lowest stone-free rate (Fig. 4A). The funnel plot showed symmetry, suggesting no publication bias between positional comparisons. All the details along with each specific net diagram were shown in the Supplements (Supplementary Figure S4, Supplemental Digital Content 3, http://links.lww.com/JS9/C16).

Figure 3.

Results of network meta-analysis: bold font indicates significant difference.

Figure 4.

Ranking probability of the outcomes in five positions.

Assessment of safety

The sample included in the assessment of complications are the same as the stone removal rate. There was also no significant difference in complication rates between the five positions (Fig. 3B). Furthermore, probability ranking was conducted, which revealed that complications were more likely to occur in the prone position > oblique supine position > supine position > split-leg position > flank position (Fig. 4B). No obvious publication bias was identified.

Secondary outcome evaluation

Due to the lack of sufficient data from two studies^27,33, comparison of operative time was generated from the other 15 RCTs^{19–26,28–32,34,35}. The SLP demonstrated significant superiority over the flank, oblique supine and prone positions [FP vs SLP (MD 41.65, 11.75–71.05); OSP vs SLP (MD 28.97, 4.25–53.98); PP vs SLP (MD 34.94, 13.72–55.27)]. The supine position also outperformed the flank and prone positions significantly [FP vs SP (MD 26.66, 3.06–50.27); PP vs SP (MD 19.79, 8.46–31.91)] (Fig. 3C). The probability ranking also illustrated that the operative time was shorter when the split-leg position was used (Fig. 4C). The operative time outcome appeared more symmetric in funnel plot, indicating a lower likelihood of publication bias (Supplementary Figure S4, Supplemental Digital Content 3, http://links.lww.com/JS9/C16).

As for hospital stay, a total of 16 studies were included in the comparison^{19–26,28–35}. The results showed that the split-leg position has a shorter hospitalization time than the other positions, as follows: FP vs SLP (MD 1.36, 0.44–2.36), OSP vs SLP (MD 0.84, 0.05–1.58), PP vs SLP (MD 0.82, 0.18–1.47), SLP vs PP (MD −0.81, −1.54 to −0.08] (Fig. 3D). The ranking of other positions in terms of hospitalization duration showed in the Figure 4D.

There were six included articles missing hemoglobin loss index^{19,22,27,28,33,34}. SLP was found to cause significantly less hemoglobin loss than OSP, PP, and SP [OSP vs SLP (MD 0.65, 0.08–1.18), PP vs SLP (MD 0.54, 0.10–0.98), SLP vs SP (MD −0.53, −1.01 to −0.04)] (Fig. 3E). The rank diagram displayed that the SLP caused the least amount of hemoglobin loss, while the OSP resulted in the highest amount (Fig. 4E). Publication bias was found in operative time and hemoglobin loss, and the stability of the conclusions needs to be strengthened.

Benefit-risk analysis and ranking

Based on a benefit-risk analysis that used the SMAA model and considered all outcomes, the split-leg oblique supine/flank position was identified as the optimal body position to balance safety and efficacy in PCNL. Combined with the above results, the higher rate of stone-clearing and lower complications make the supine and reclining supine positions more advantageous than prone positions (Fig. 5).

Figure 5.

The result of benefit-risk assessment utilizing the SMMA model.

We conducted a Grading of Recommendations Assessment for the outcomes. The evidence levels for the outcomes related to stone-free rates and complications were both recommended as Moderate (Supplementary Table S6, Supplemental Digital Content 3, http://links.lww.com/JS9/C16).

Discussion

In our NMA, we consider that the prone, oblique supine and flank positions exhibit suboptimal performance, while the supine position and split-leg position are deemed more efficacious and safer. While no statistically significant differences were observed in the primary outcome measures, we still noted that the split-leg position reduced surgical duration, hospitalization duration, and hemoglobin loss.

Several modified postures, such as the Galdakao-modified Valdivia, the modified supine Double-S, and the Barts-modified, have been created since the introduction of the supine position³⁶. The emergence of diverse and intricate new PCNL positions has also posed challenges in conducting direct meta-analyses. We used NMA to form a network comparison system to fill this gap by comparing direct and indirect comparisons. Drawing definitive conclusions regarding the superiority of multiple positions remained difficult.

Our NMA found that the split-leg and supine postures were associated with a higher stone-free rate than the prone position. This run counter to the findings of two previous meta-analyses^37,38 but aligns with a meta-analysis which exclusively included RCTs³. We place greater confidence in the reliability of our conclusions due to the higher quality of evidence from RCTs. The supine position with a horizontally or slightly downward sloping percutaneous tract and lower pressure in the pelvis facilitates spontaneous expulsion of stone fragments during the procedure, leading to a higher rate of stone clearance³⁹. A ureteroscope and the recently suggested ECIRS method can be used in the supine position for the therapy of complicated stones, obviating the need for numerous percutaneous approaches and allowing for a more thorough investigation of the renal calyces. All of the above enabled the development of improved stone clearance rates⁴⁰.

Based on our research, we concluded that the prone position was associated with a greater risk of complications. These findings may be attributed to the limitations inherent in the prone position itself. Flank, oblique supine, supine, and split-leg positions are preferable to prone position as they ease dyspnea caused by chest and abdominal compression, especially in patients with heart disease and obesity⁴¹. These positions also reduce challenges faced by anesthesiologists during surgery, along with helping doctors to manage airways efficiently⁴². This would assist in reducing the incidence of surgical complications, reducing the risk of injury to patients, and require fewer healthcare workers. Moreover, colonic injuries occur more frequently in the prone position PCNL^43,44. Supine patients have a decreased risk of colon damage than prone patients, which can help prevent potentially life-threatening consequences^6,45.

Consistent with prior meta-analyses, we found that hospital stays were roughly same whether the patient was lying supine, oblique, or prone^3,37,38. However, split-leg has the shortest length of hospital stay and may be associated with fewer complication rates. Faster procedure times reduce the risk of febrile infection in later stages, thereby reducing length of hospital stay. At the same time, the length of hospital stay is more affected by the local medical insurance system and the social health system.

The hemoglobin loss during surgery may be influenced by physiological changes in kidney perfusion and venous return, which are affected by patient positioning. While 0.8–10% of instances of acute bleeding needing blood transfusion have been observed during the prone position⁴⁶, investigations have shown varying degrees of hemoglobin loss when patients were lying on their backs. The spilt-leg oblique supine/flank position minimized bleeding during the PCNL procedure due to its ability to relax the limbs and increase surgical space, allowing for easier instrument insertion, thus reducing unnecessary repetitive movements and minimizing intraoperative movement and vibration.

Our NMA concluded that selecting the split-leg oblique supine/flank position would lead to superior clinical outcomes, which might contribute to the reduced operative time, length of hospital stays, and hemoglobin loss while maintaining stone-free rates and keeping the rate of complications low. The split-leg position is a modified supine position, with the inclination angle located between fully lateral and fully supine, and the upper limbs crossing on one side, similar to a riding position⁸. Through our review of the picture of the position in Al-Dessoukey’s article, we included the article in the SLP group while we found that the oblique supine position was actually the split-leg position²⁹. This position can better allow the lumbar spine to extend straight and provide more space for puncture and assistance in opening the renal angle, thus better accessing the lower pole of the kidney and enabling a larger area for renal endoscopy. In Ahmed AF’s study, SLP had a longer tract formation time than PP, which may be due to the learning curve of the surgeons⁴⁷. The shorter total procedure time may be also attributed to the fact that the patient did not need to be repositioned and resterilized after changing position and the surgeon did not need to scrub and dress again²⁴.

Our analysis has significant implications for optimizing PCNL positions. We advocate for an individualized approach to patient positioning, taking into account factors such as calculi location, anatomical variances, and comorbidities. Overweight and obesity among the adult patients included in this study may have had a major bearing on the findings. Therefore, split-leg and supine position were more applicable to the obese population. Our findings could guide preoperative discussions, enabling surgeons to weigh the merits of various positioning strategies with patients, fostering informed decision-making. Furthermore, learning curve of doctors was crucial for the success of the operation and the health of the patients⁴⁷. Integrating the insights into surgical training programs could sharpen the skills of emerging urologists.

There are certain limitations to our NMA. It was challenging for the included RCT trials to completely blind participants, which might introduce bias into the results. Additionally, the small number of studies on split-leg and flank positions included might result in significant statistical bias and errors. High heterogeneity was found only in continuous variables, which may be due to the large differences in measures of duration of surgery and length of hospital stay between different regions and hospitals. Objective measures such as stone clearance rate and complications were not significantly heterogeneous across studies. The inconsistency in the utilization of laser, ultrasonic, and ballistic techniques for PCNL surgery across the reviewed literature may also exert a certain influence on the outcomes. Furthermore, different admission criteria for stones in different literature and the use of ultrasound or lens guidance, among other factors might be one of the factors contributing to heterogeneity. To better evaluate the safety and effectiveness of these surgical positions, more high-quality clinical studies are needed.

Conclusion

The effectiveness and safety of all positions are recognized. The supine and oblique supine positions exhibited more advantages than prone positions. The split-leg position might be a more effective choice considering high SFR, less operative and hospital stay time, and fewer hemoglobin loss. Experienced surgeons should choose the position based on the specific conditions of the patients and their proficiency.

Ethical approval

The study is a meta-analysis, so ethical approval is not applicable.

Sources of funding

The study was supported by the Foundation of Sichuan Provincial Department of Science and Technology (2023YFS0029, 2022YFS0304).

Author contribution

P.L., Y.M., and B.L.: conceptualized and designed the investigation; P.L., L.X., and X.J.: data collection; P.L., B.L., and Y.M.: data analysis and/or interpretation; P.L. and Y.M.: drafted and critically revised the manuscript; H.L. and K.W.: reviewed and authorized the final version of the manuscript.

Conflicts of interest disclosure

All authors have no conflicts of interest or financial ties to disclose.

Research registration unique identification number (UIN)

1. Name of the registry: PROSPERO.

2. Unique identifying number or registration ID: CRD42023409385.

3. Hyperlink to your specific registration (must be publicly accessible and will be checked): https://www.crd.york.ac.uk/prospero/display_record.php?ID=CRD42023409385.

Guarantor

P.L., Y.M., B.L., X.J., L.X., H.L., and K.W.

Data availability statement

The study is a meta-analysis, so the data statement is not available. Comparison of Safety and Efficacy of Different Positions in Percutaneous Nephrolithotomy: A Network Meta-analysis.

Data collaboration statement

This article and its supplementary information files contain all data generated or analyzed during the course of this investigation.

Provenance and peer review

Not commissioned, externally peer-reviewed.