Introduction

Percutaneous nephrolithotomy (PCNL) is the treatment of choice for large renal stones and has a 78–95% of success rate. However, the overall surgical complication rate of PCNL is as high as 83%, with a major complication rate of 1.1–7% [1]. Colon injury during PCNL is classified as a grade III complication according to Clavien–Dindo classification [2]. Its incidence is only 0.2–0.8%, which is very low [3]. However, it is a serious complication that can be fatal and difficult to predict preoperatively.

Posterolateral colon (PLC) and retrorenal colon (RRC) are the significant risk factors of colon injury during PCNL [2,3,4,5]. The incidence of RRC varies from 1.7 to 10% [6,7,8]. The incidence of PLC ranges from 3 to 19% [5, 9]. Routine preoperative CT scan were shown to be necessary and cost-effective for imaging renal stones and determining the presence of RRCs in the previous studies [10, 11]. In general, the operative procedure of PCNL is performed in the prone position. However, preoperative computed tomography (CT) is carried out in the supine [12, 13]. The effect of colon’s movement around the kidney according to changes of the patient’s position on potential colon injury during the initial percutaneous puncture has not been fully established yet. The incidence of PLC and RRC is different according to each position. A few studies have compared the incidence of PLC or RRC according to different positions in the same person and only studies with few cases have been reported [6, 9, 14]. However, no study has reported the incidence of PLC and RRC in both prone and supine positions. The purpose of our present study was to investigate the tendency of changes in the anatomical relationship between the kidney and colon in the prone and supine positions to evaluate their potential effect regarding colonic injury in the percutaneous approach used for PCNL.

Materials and methods

A total of 113 patients with renal stones requiring PCNLs from February 2017 to May 2018 were enrolled, and all patients underwent abdominal CT scans in the supine (preoperative) and prone (postoperative) positions. During PCNL, percutaneous puncture was performed under fluoroscopy in the prone position. Preoperative CT scan was carried out on the day before surgery. Postoperative CT scan was performed postoperative day 1 of PCNL to confirm the presence of residual stones. Therefore, there was no risk of additional exposure to radiation for the study. Patients of both males and females were included in the study.

Exclusion criteria

The following patients were excluded because they may have significant changes in anatomical relationships: patients < 18 years old and patients with congenital anomalies (ectopic or horseshoe kidney), large exophytic renal cysts, huge perirenal hematoma, history of previous bowel surgery or renal surgery (except PCNL), upper abdominal space-occupying lesions, ascites, severe kyphoscoliosis, spinal deformities, and small kidneys. Two experienced urologists reviewed all CT scans.

This study was approved by the local Institutional Review Board (IRB) of the Catholic University of Korea (approval number: KC18RESI0759). The IRB waived the informed consent, due to the retrospective nature of the study.

Definitions of retrorenal and posterolateral colons

The definitions of PLC and RRC have been provided in several studies previously [14,15,16]. In the present study, similar methods were applied to define PLC and RRC, by drawing virtual horizontal lines through objective indicators.

In the transverse view of CT images, three virtual parallel lines were drawn horizontally. First line along the anterior surface of the left kidney, the second one along the posterior surface of the left kidney, and the other line passing the posterior calyceal border. These three horizontal lines defined four zones of the space which is surrounding the kidney. Zone 1 was the anterorenal zone; Zone 2, the anterolateral; Zone 3, the posterolateral; and Zone 4, the retrorenal. On each slice, position of the posterior margin of the colon was an indicator for classification (Fig. 1).

Fig. 1
figure 1

Diagram showing the position of the colon surrounding the kidney. Three virtual parallel lines were drawn horizontally. First line along the anterior surface of the left kidney, second one along the posterior surface of the left kidney, and the other one passing the posterior calyceal border. (1) anterior zone; (2) anterolateral zone; (3) posterolateral zone; (4) retrorenal zone. Lower left, posterolateral colon (PLC, indicated); Lower right, retrorenal colon (RRC, indicated)

Full size image

The transverse view of the CT image was set at each posterior calyx in the upper calyces, mid-renal hilum level, and lower calyces. The data were obtained from the upper, middle, and lower poles of both sides of the kidneys in the supine and prone positions.

The skin-to-kidney distance, subcutaneous fat thickness, and perirenal fat thickness were measured by the following method. An imaginary anterior–posterior line at the point of the posterior margin of the calyx was drawn and the vertical distance through this line was measured to find the quantified factors.

Statistical analysis

The data were analyzed using IBM SPSS version 22.0 (IBM, Chicago, Illinois, USA). Values are presented as mean ± standard deviation. The groups were compared using χ2 analysis, paired t-tests, and repeated measures analysis of variance. p value was considered statistically significant if less than 0.05. Kruskal–Wallis test was used for comparative analysis of colon movements between supine vs. prone position. χ2 analysis and paired t-tests were used in univariate analysis. One-way ANOVA and Kruskal–Wallis test were used in multivariate analysis for evaluating risk factors of PLC and RRCs in the prone position.

Results

Patient demographics

Among 113 patients initially collected, 11 patients were excluded. Finally, 102 patients (14 conventional PCNL patients and 88 tubeless PCNL patients) were enrolled. All participated patients underwent CT scans in both supine and prone position. Among all cases, 42 were females and 60 were males, with a mean age of 54.2 years (range from 23 to 83). Mean body mass index (BMI) of the patients was 24.6 (kg/m2).

Supine vs. prone position results

PLC and RRCs were found in 16 (15.7%) cases in the supine position, and 25 (24.5%) in the prone, respectively (Fig. 2). Twelve patients had PLC or RRCs in both positions.

Fig. 2
figure 2

Incidence of posterolateral colon (PRC) and retrorenal colons (RRC) in the supine and prone position. Left; incidence in the supine position. Right; incidence in the prone position. The same location on the upper and lower calyces of the left kidney was significantly different

Full size image

Five cases in the supine position and eight cases in the prone showed PLC or RRCs at more than one site. Four cases in the supine position and six cases in the prone had bilateral PLC or RRCs. Compared to the supine position, the incidence of PLC or RRCs was significantly higher in the left upper pole (p < 0.001), and the left lower pole of the kidney in the prone position (p = 0.005).

Comparative analysis was conducted on the position of the colon at the same level during changes of position from supine to prone. Overall, the relative anterior movement of the colon was more prominent on the left kidney than on the right one (Table 1).

Table 1 Tendency in colonic movement by change in position from supine to prone (n = 102)
Full size table

Risk factors of PLC and RRCs in the prone position

Quantitative analysis was conducted to determine the risk factors of PLC and RRCs in the prone position (Table 2). Age (p = 0.004), BMI over 30 kg/m2 (p = 0.028), and thickness of the perirenal fat layer (p = 0.048) were significant risk factors of PLC and RRCs in univariate analysis.

Table 2 Univariate and multivariate analysis of risk factors associated with posterolateral and retrorenal colons in the prone position
Full size table

Age (OR: 1.073, p = 0.002), BMI over 30 kg/m2 (OR: 0.019, p = 0.028), and thickness of the perirenal fat layer (OR: 0.892, p = 0.047) remained significant factors in the multivariate analysis.

Old age (> 68 years) increased the risk of PLC and RRCs. However, thicker perirenal fat thickness and high BMI (over 30 kg/m2) were in negative relation with the risk.

Discussion

PLC and RRCs are the significant risk factors of colon injury during PCNL [3,4,5]. According to a report by the endourological Society’s Clinical Research Office, 80% of the PCNLs worldwide are performed in prone positions [13]. However, preoperative CT scans are carried out in the supine position in most cases, even if the operation procedure is performed in the prone. Changes in the relative location of the kidney and other adjacent organs due to position changes were not fully established yet. Furthermore, perioperative complications, such as colon injury, can also occur even if the operation is performed by experienced practitioners. Inappropriate surgical planning can be the cause of such complications. Therefore, knowledge about the retroperitoneal space anatomy and its variations using CT scans are important, especially regarding the relative anatomical location of the colon and kidney according to the patient’s position.

Despite the significance of PLC and RRCs in PCNL, their incidence varies considerably in published reports. The present study revealed that PLC and RRCs were present in 16 (15.7%) cases in the supine position, and 25 (24.5%) cases in the prone. These results were not significantly different from those reported in other studies on RRCs (supine 1.7–10%, prone 6.8–20%) and PLCs (3–19%) [5,6,7,8,9]. The prone position is associated with posteriorly located colons. The incidence of PLC and RRCs was suggested to be higher in the prone than in the supine, despite no statistically analyzed study [9, 17, 18]. Reddy et al. [11] reported 4 cases (1.14%) of RRC, in which 2 (0.76%) were observed on the left kidney, 1 (0.38%) on the right kidney and 1 (0.38%) with bilateral RRC among 262 patients with CT scan before PCNL. One remarkable result of the present study was that the incidence of PLC and RRCs was different significantly between the two positions. Compared to the supine position, the incidence of PLC and RRCs was significantly higher especially in the left kidney in the prone position, at the upper pole (p < 0.001) and the lower pole (p = 0.005), respectively. In the article reported by Tuttle et al., risk of organ injury during PCNL was significantly increased for lower pole kidneys in the prone position (15.1% in prone vs. 6% in the supine) [18]. This may be a result of the increased occurrence of PLC and RRCs in the prone position which was compared to supine, especially at the kidney lower pole level. Boon and colleagues evaluated 333 cases of CT scans in the supine position only and demonstrated that the left side colon was related with the risk of potential injury in 16.0% of the total cases, whereas the right colon was at risk of potential injury in 9.0% of the cases [19]. Our study showed similar results in that the risk of potential colon injury was present in both supine and prone positions.

However, studies on factors that can increase the incidence of PLC and RRCs are lacking. Therefore, risk factors of PLC and RRCs in the prone and supine position were investigated regarding colon injury during PCNL.

To the best of our knowledge, our present study was conducted by enrolling the largest numbers of subjects to date who underwent CT scans in both supine and prone positions to investigate the risk factors of PLC and RRCs. Our results were not largely different from those of previous studies, which showed that the risk factors of colon injury during PCNL were old age, RRC, gender (female), underweight status, decreased retroperitoneal fat tissue, and previous abdominal surgery [3, 6]. These results suggest that changes in the relative position of an organ according to patient’s position might affect the planned percutaneous access and are related to a higher incidence of colon injury during PCNL. In this study, we measured the severity of obesity based on the thickness of the perirenal fat tissue using CT scans, in addition to BMI. The thickness of perirenal fat tissue is an indicator of obesity because it affects organ location. It can be a more objective and personalized indicator than BMI. Therefore, it is important that this was a statistically significant risk factor.

It is speculated that thinner perinephric fat layer gives more perirenal space for the colon to move to posterior or lateral side of the kidney in the prone position. According to Hadar and Gadoth et al. [20] colons tend to move anterior to the kidney in persons with thicker perinephric fat, whereas colons tend to be posterior and lateral to the kidney in persons with thinner perinephric fat. They also reported that colons have a tendency to be placed posterior and lateral to the kidney in patients over 60 years old. In elderly people, the amount of perinephric fat tissue decreases, which results in the movement of the colon posterior and lateral to the kidney.

Currently, the mechanism by which PLC and RRCs occur is unclear. One of the hypotheses for changes in location is the embryological theory [16]. Until the 12th week of fetal development, descending colon moves around the lateral border of the left kidney. From the 12th to 17th weeks, left mesocolon is fused with the parietal peritoneum which is attached to the left renal surface. This causes adhesion of the colon with lateral space of the kidney and moves the colon forward. The degree of adhesion leads to variations in colonic position. A theory proposed by Raptopoulos [21] is that the junction of the anterior renal fascia and laterocolonal fascia that determines the posterior margin of the colon is attached by thin fibers that can be separated easily. Thus, potential space can be created to affect the colon’s posterior movement. In a study of 140 patients, Hadar et al. [20] found that patients aged from 0 to 19 years old do not have posterior colon displacement, showing an increased incidence of displacement with increasing age. Prassopoulos et al. [9] also showed that the perirenal fat accumulation was greater in males than in females. They hypothesized that there was a difference in visceral fat accumulation according to age and gender of patients.

By enrolling the largest number of subjects to date who underwent CT scans in both positions, changes in the anatomical relationship of the visceral organs according to the patient’s position were established objectively in the present study. Upper and lower pole of the left kidney was the location where the occurrence of PLC and RRC increased significantly when position change was done from supine to prone. When the patient’s position was changed from supine to prone, posterior movement of the colon was observed prominently on the left side.

Conclusions

Initial percutaneous approach in PCNL should be carefully processed to avoid colon injury, when operating the patients with high risks of PLC and RRC, such as old age, lower BMI less than 30 kg/m2, and a thinner layer of perirenal fat. PLC and RRCs were observed more frequently at upper and lower calyces of the left kidney in the prone position when it was compared to the supine.