The PCNL is well established as the first-line treatment for complex and high volume upper urinary stones, with stone clearance rates ranging from 60 to 90% [16, 17]. The ultimate goal of the operation is to reach a stone-free status with minimal morbidity. However, Just as all other surgical interventions, PCNL confers differing risks of complications and residual stone. Preoperative stratification of risk factors and reliable surgical planning remain of utmost priority for patients and urologists, particularly in the context of weighing the benefits of operation against potential risks and adverse effects. More and more scholars have taken advantage of perioperative factors to predict SFS and risk of complications after PCNL. To date, multiple attempts have been made to elaborate a scoring system that benefits from patients. However, none of the proposed scoring system have been generally adopted as a standard owing to their limited predictive accuracy, lack of validation and clinical utility.
In a previous study, Chinese scholars proposed SHA.LIN stone score for assessing the stone-free rate of PCNL and investigated the clinical value in patients undergoing PCNL. However, unlike Guy’s or S.T.O.N.E. scoring system, SHA.LIN score has not been universally known to urologists and is still pending validation. This study aimed to introduce the SHA.LIN score and compare the accuracy to S.T.O.N.E. and Guy’s stone scores in predicting postoperative outcomes. Compared to previously proposed stone scoring systems, the SHA.LIN stone score uses variables that are easily calculated from NECT and do not require specialized software. The stone score variables were defined based on operation experience and draws on extensive literature reviews and existing stone scoring systems [5–7].
The predictive accuracy of Guy’s and S.T.O.N.E. scoring systems has been summarized and compared in published works. Study results have varied, however. As we know, Guy’s score is reproducible and is simple to apply in routine clinical practice for assessing surgical risk. However, it does not account for critical variables such as stone burden, calyceal involvement and stone density [6]. Most research reported that these parameters have an important influence on postoperative outcomes [10, 15, 18]. In addition, Guy’s score has four grades limiting the ability to evaluate the complexity of stone characteristics. Although S.T.O.N.E. and SHA.LIN scores have common parameters, such as stone burden, tract length, degree of hydronephrosis, and stone essence, the definitions of these parameters are still different [5]. For instance, stone burden is an essential parameter in two scoring systems, whereas in SHA.LIN score, stone burden was estimated by combining stone length and maximum length in CT slice in square millimeters. If stone is multiple, the SHA.LIN score calculate the sum of every stone area. The S.T.O.N.E. score only calculated the largest stone by combining the measures of length and width in square millimeters. We believe that the stone burden of SHA.LIN can better reflect the complexity of stone characteristic. In the S.T.O.N.E. score, the definition of calyx and imaging plane is not standardized. Stone size/number of calices involved is also not standardized and is variable between different observers [8, 9]. Hydronephrosis degree score is subjective and does not have a clear definition in S.T.O.N.E. score [8]. In SHA.LIN score, the author not only refined the number of calices containing stones, anatomic distribution of stones and number of involved calices but also made a clear definition of each variable. Thus, urologists can perform a standardized evaluation of every patient with a CT scan, increasing the reliability of the outcome assessments.
In the present study, stone clearance with PCNL was 69.3%, whereas similar studies by Krishnendu[15] had SFS of 71.5%, Thomas[19] had SFS of 62.0%, Labadie[20] had SFS of 56.0%. Stone burden is the most crucial variable for predicting the SFS. In our study, there was a statistically significant difference in the mean size of stones in the two groups (p < 0.001). The presence of stones in multiple calyces was significantly associated with a decreased stone-free rate in comparison to single calyceal involvement. Staghorn stone had a significant association with SFS, with partial staghorn stone had 56.7% and complete staghorn stone had 48.5% SFS. Labadie[20] reported that staghorn renal stone had shown 40% SFR among operated patients. In Guy’s score, partial staghorn as Grade III and complete staghorn as Grade IV, stone clearance rates were 35% and 25% respectively [6]. Research showed that stone distribution and location have an essential on SFS. There are two opposing opinions in the determination of stone distribution in S.T.O.N.E and Guy’s scores [8]. The Guy’s score similarly assigns categorizations according to anatomic distribution in the renal pelvis, lower calyx, middle calyx, and upper calyx. In contrast, S.T.O.N.E. score prioritizes the number of stones involved calyces, with an overall algorithm determining how much weight each location contributes to complication and SFS [8]. Whereas the SHA.LIN score was referred to the above two scoring systems. The authors not only considered the effect of staghorn stones on the stone clearance rate, but also redefined the distribution of stones in renal calyx. Stone in renal pelvis or mid/lower calyx is assigned 1 point. Stone in the upper calyx is assigned 2 point. Stones in a calyceal diverticulum or partial staghorn stone is assigned 3 point. Full Staghorn stone is assigned 4 point. We believe this classification method is more indicative of the complexity of stone characteristics. In the present study, three stone scores were significantly associated with SFS and operation time. These conclusions are consistent with previous reports [10, 21, 22]. We noted the comparable accuracy of the SHA.LIN score (AUC, 0.829), Guy’s score (AUC, 0.731), and the S.T.O.N.E. score (AUC, 0.789) for predicting SFS. This can be interpreted as SHA.LIN had a higher power to predict the SFS after PCNL than the other two scoring systems.
Bleeding is one of the most unpredictable and threatening complications during PCNL. Published data showed that stone burden, degree of hydronephrosis and staghorn stone are associated with an increased risk of blood loss complication [23]. The variables of the SHA.LIN score include these risk factors. In recent studies, the relationship between EBL and stone scores was unclear [13, 15]. Akhavein reported that S.T.O.N.E. score had significant correlation with SFS, but did not find correlation with EBL in a study of 117 patients [24]. In previous study of 437 patients, they found there hadn’t significant correlation between S.T.O.N.E. score and EBL [13]. Labadie, in a study of 246 patients who underwent PCNL, concluded that Guy’s and S.T.O.N.E. scores had significant correlation with EBL [25]. The differences in these studies may be due to a low number of renal stone patients or poor universality of the scoring system. In the present study, EBL and haemoglobin change were significantly correlated with SHA.LIN stone score. ROC analysis area AUC demonstrated a more accurate prediction of blood loss based on the SHA.LIN stone score in comparison with the other two stone scores. However, the conclusion needs further validation. As a new scoring for stone characteristics, SHA.LIN can accurately predict SFS and can be used to assess surgical risk factors. However, it has limitations, including lack of accurate measurement of stone size and individual surgical experience. If these issues are resolved, the SHA.LIN stone score could be a more precise predictive tool for postoperative outcomes.
There are still some limitations in the present study. Firstly, this is a single center, small sample size and retrospective study. To solve these problems, we standardized clinical data collection and analysis procedures and used rigid parameter definitions. In China, SHA.LIN stone score is the first proposal of a predictive method for the SFS after PCNL. We only compare it to Guy’s and S.T.O.N.E. scores, not to Clinical Research Office the Endourological Society nephrolithometric nomogram. Further research should be performed to validate the results. Lastly, we did not use CT to detect the SFS randomly in all patients like other studies[5, 20] had done, which may cause bias in SFS calculation. To solve this problem, all patients received KUB and ultrasound every month for three months to detect SFS.