Comparing Thulium Fiber Versus High-Power Holmium Laser Lithotripsy Combined with the Flexible and Navigable Suction Access Sheath in Flexible Ureteroscopy for Kidney Stone Disease: A Propensity Score Matched Analysis by the Global FANS Collaborative Group

Introduction

Currently, thulium fiber laser (TFL) and holmium:yttrium-aluminum-garnet laser (HL) are the two strongest contenders used both in adults ¹ and children ² for laser lithotripsy (LL) in retrograde intrarenal surgery (RIRS). The former appears to have an edge in terms of ablative efficiency, ³ even if a recent randomized trial showed no significant clinical advantage of TFL when compared with pulse-modulated HL for both ureteral and renal stones. ⁴

The onus is on the surgeon to exercise caution to prevent laser-induced damage ³ and use the laser responsibly enough to perform efficacious lithotripsy. ⁵ To maximize stone-free rate (SFR), surgeons have a choice of using different LL techniques or modulating different power and frequency settings, ⁶ which if not used carefully can counterproductively lead to ureteral thermal damage. ^7,8

Ureteral access sheath (UAS) also plays an important part in improving the clearance of fragments and reducing intrarenal pressure. When combined with suction, UAS plays an integral role in reducing intrarenal pressure and temperature ⁹ and improving SFR, ¹⁰ making flexible ureteroscopy (F-URS) safe. The flexible and navigable suction access sheath (FANS) with a unique ability of having a bendable proximal tip that can be directed to different calyces has shown much promise. ¹¹ A recent study showed that for renal stones in adults with normal renal anatomy, FANS improves SFR, minimizes complications, and provides a possibility of zero residual fragment (RF). ¹² Literature is lacking on what are the common laser settings used by surgeons and their influence on SFR for different stone size when using these powerful lasers.

In lieu of the evidence gap, the primary aim of this study is to compare the perioperative outcomes of LL with TFL or high-power HL (HPHL) in adult patients with kidney stone(s) undergoing RIRS using FANS. The secondary outcome is to report surgeons’ preferred techniques and laser settings.

Material and Methods

The global FANS collaborative group is a worldwide multicenter collaborative effort to study the use of FANS in RIRS in real-world practice for renal stones in adults (>18 years) with normal renal anatomy. Twenty-five centers prospectively contributed to data in 394 patients in whom F-URS was done with FANS from August 2023 to January 2024. Study protocol and 30-day outcomes of the full series were already published. ¹² Only patients in whom a FANS was successfully deployed on the initial attempt were enrolled, irrespective of whether they had a preoperative stent. For this analysis, we focused on patients who had RIRS using HPHL (Lumenis Pulse™ 100H, Boston Scientific, Marlborough, MA) or TFL (Fiber Dust, Quanta System, Varese, Italy or Soltive, Olympus, Hamburg, Germany or Urolase SP 60W, IPG Photonics, Oxford, MA) to analyze the influence of the two lasers on the outcomes. Institutional review board approval was obtained by an ethical board-approved FANS registry curated by the Asian Institute of Nephro-Urology, Hyderabad, (#AINU 12/2022).

The size and type of ureteroscope were determined by the surgeons’ preferences and the available resources. The choice of energy source for RIRS and perioperative decisions such as prestenting and postoperative exit strategy was at the respective surgeons’ discretion. All patients had a preoperative and at least one postoperative non-contrast computed tomography (CT) scan to assess stone features and RFs within 30 days of operation. Children and patients who had abnormal renal anatomy, ureteral stones, or insufficient data records were excluded.

Details about the type of laser, preferred technique, and lasing setting and mode of lithotripsy were recorded. The ease of use of FANS with the two lasers was documented by using a 5-point Likert-type scale (1 = excellent; 2 = very good; 3 = good; 4 = average; 5 = difficult). Ureteroscopy time was the time from FANS placement to its removal. Lasing time was the time recorded as obtained from the machine display. Total surgical time was defined as the time from the start of cystoscopy to exit strategy (stent, ureteral catheter, or nil drainage). Loin pain was evaluated on the first postoperative day during the hospital stay for inpatients or by a phone call for those patients who underwent operation on a same-day discharge basis, using a standard 10-point visual analogue scale, with 0 representing the lowest score.

Stone-free status was recorded by urologists and graded as:

Grade A: 100% stone-free status. No fragments or dust were visible on the CT scan.

Other outcomes of interest included postoperative complications, readmission, and reintervention within 30 days.

Continuous variables are described using median and 25th–75th quartiles, whereas categorical variables are described using absolute numbers and percentages. Differences between the groups were compared using the χ² test or Fisher exact test for categorical parameters and the Mann–Whitney U test for continuous variables.

A propensity score matching (PSM) was used to reduce confounding in the baseline characteristics. Propensity scores were calculated using a logistic regression model, and one-to-one nearest-neighbor matching for age, gender, stone location (upper pole, middle pole, lower pole, or multiple locations), stone volume, and Hounsfield unit. To ensure optimal matching of covariates, the caliper width was started at 0.2 ^13–14 and decreased in increments of 0.01 until the absolute standardized mean difference (ASMD) for all covariates was <0.1, ¹³ which was achieved at a caliper width of 0.06. All statistical comparisons were then repeated for the matched cohort like the overall cohort.

A multivariable logistic regression analysis was performed in the matched population to evaluate factors associated with Grade A SFR. Variables included in the regression model were selected a priori based on data that have been suggested in previous literature to impact SFR (stone location, stone volume, Hounsfield unit, and operative times). Data are described using odds ratio (OR), 95% coincidence interval (CI), and p value. All statistical tests were performed using R Statistical language, version 4.3.0 (R Foundation for Statistical Computing, Vienna, Austria) with p < 0.05 indicating statistical significance.

Results

Table 1 shows patients’ baseline characteristics in the unmatched and matched cohort. For the unmatched cohort, patients in the HPHL group were significantly older compared with those in the TFL group. There was a significantly higher proportion of men in the HPHL group (61.4% vs 53.6, ASMD = 0.16). Median stone volume was significantly larger in the TFL group [1440 [784, 2112] vs 1112 [708, 1500] mm³, ASMD = 0.32], with a higher prevalence of stones located in the middle pole (43.1% vs 29.8%, ASMD = 0.32). Median stone density was also significantly higher in the TFL group [1440 [784, 2112] vs 1112 [708, 1500] Hounsfield unit, ASMD = 0.32].

After PSM, 96 patients from each group were included. The baseline demographics and stone characteristics were comparable in the matched cohorts (all covariates were matched to an ASMD of <0.1) (Table 1). Table 2 shows the intraoperative characteristics of unmatched and matched populations. There was a significant difference in sheath size, where 11–13 Fr was the most prevalent in the HPHL group (58.3%), whereas 10–12F was in the TFL group (44.8%). There was no difference between the groups regarding the use of disposable scopes, lithotripsy mode, ureteroscopy, lasing, and total surgical time. In the matched cohort, overall median frequency and energy for dusting only was similar for both lasers, but there was a significant difference in overall energy deployed notably higher in HPHL [1.2 (1.0, 1.5) J vs 0.8 (0.6, 1.0) J in TFL, p < 0.001] and a higher frequency setting for TFL compared with HPHL [20 (20, 35) Hz in TFL vs 15 (10, 20) Hz, p < 0.001]. No issues with suction occurred because blocking fragments were noted in any case. Exit strategy after lithotripsy did not differ significantly. There was a significantly better performance of FANS in terms of manipulation (1 [1, 2] vs 2 [2, 3], p < 0.01) and visibility (1 [1, 2] vs 3 [1, 4], p < 0.01) in the TFL group.

Table 3 presents the preferred laser setting correlation for stone diameter and volume in the PSM cohort. Laser settings were similar for both groups when using TFL and HPHL even as the stone density increased when used both selectively in dusting mode and fragmentation mode. The differences noted in laser setting preference show that for stone <1 cm, for fragmentation, the total power deployed for TFL (26.4 W vs 10 W) was higher. However, as this crossed 2 cm, surgeons reportedly used much higher total median power in the HPHL group (18 W vs 16 W). Irrespective of stone volume, the median total power deployed for dusting was the same for the HPHL group (15–16 W) and no change for fragmentation (18 W). When using the TFL, the energy power settings increase proportionately as stone volume increases for dusting (13.2 W, 14.6 W, and 16.25 W). However, when stone volume was >3000 mm³, the median power needed for fragmentation was much lesser (26.4 W vs 21 W).

Postoperatively (Table 4), there was no case of sepsis, and blood transfusion was necessary for only one patient in the HPHL group (Clavien grade 2). The incidence of transient fever (Clavien grade 2) was low and similar between the groups (3.4% in HPHL vs 3.8% in TFL group, p = 0.61). Postoperative day one loin pain did not differ significantly [2 [1, 3] in HPHL vs 2 [1, 2] in TFL group, p = 0.61]. The postoperative stay was similar in both groups [1 [0, 2] in HPHL vs 1 [0, 1] days in TFL group, p = 0.12]. Grade A SFR on inspection at the end of lithotripsy was reportedly higher in the TFL vs HPHL group [71.9% in the TFL group vs 50% in the HPHL group, p < 0.01]. Thirty-day Grade A SFR did not differ significantly between the groups with 52.1% for HPHL vs 64.6% for TFL (p = 0.11). Overall, reinterventions were low with one patient in the HPHL and three patients in the TFL group planned for ancillary treatments.

Multivariable logistic regression analysis of the matched cohort shows that the use of TFL was significantly associated with higher odds of achieving Grade A stone-free status (OR 1.95, 95% CI 1.01–3.82, p = 0.04) (Table 5).

Discussion

In an initial analysis, FANS has shown the potential to be a game changing accessory for RIRS, mainly as it can achieve the trifecta of high single intervention stone-free status (i.e., zero RF), negligible minor complications, and a very low reintervention rate. ¹⁵ Yet, we need to ascertain which parameters might have actually contributed to these, considering they are much different to those reported by studies like FLEXOR. ¹⁶ Since FANS is a new concept to perform RIRS, it is useful to analyze which laser has a more effective role when performing RIRS with FANS.

The debate between TFL and HL as the best or ideal laser in F-URS remains unsettled. Uleri et al. reported in their recent meta-analysis that compared with HL, TFL had better SFR especially if the criterion is to have no RF (but this was found in pediatric studies only). ¹⁷ In adults, however, using a 3 mm cut-off for RF, no significant difference was seen when TFL was compared with MOSES technology and neither could help achieve a zero RF status post RIRS. ¹⁷ Castellani et al. showed that when TFL is used with a traditional UAS vis-a-vis HL with MOSES technology a much higher SFR is possible (85% vs 56%, p < 0.001). ¹ Our regression analysis supports this, albeit with a new generation of UAS. The possible reasons could be that when combined with suction, the already known benefits of better dusting and better ablation with TFL ¹⁸ are further improved by virtue of being able to provide better vision and better ability to simultaneously remove dust and debris from all parts of the collecting system by navigating or steering the bendable tip of the sheath in the renal collecting system. ¹⁹

As we enlisted the settings that are used in different stone sizes and stone volumes, we hope that perhaps surgeons could benefit from the value of these settings if they are using FANS with either TFL or HPHL. Interestingly, as the stone burden increases, especially considering stone volume >3000 mm³, the median overall operative time was much shorter in the TFL group (55 minutes vs 70 minutes), and this was also seen for lasing time (25 minutes vs 32 minutes). As these lasing parameters are based on surgeons’ reported settings, we can only infer that TFL might be more efficacious when dealing with larger stones, but HPHL could be equally efficacious for lower stone burdens. Also when tackling large burden stones, keeping operative time short could lessen the risk of iatrogenic and infectious complications ²⁰ and increase scope longevity, ²¹ especially when using reusable scopes. In our study, we compared the influence of LL when using two different lasers by accurately measuring not just stone density but also true stone volume. In fact, Majdalany et al. reported that stone volume, and not stone density, correlates better to lasing time, especially when using MOSES technology, ²² and stone volume may be a better predictor of SFR, especially in larger stone burden. ²³ Indeed, from our study, both lasers can help achieve a high SFR with TFL having an edge over HPHL as seen in regression analysis for achieving Grade A stone-free status.

The preferred settings and lasing techniques surgeons involved in this study used can be a reference point for users and perhaps more studies can validate the best laser settings using FANS. Interestingly, among the techniques reported for LL, none reported using popcorning. This could be because of the actual ability to remove whole fragments along with dust and hence popcorning was not needed because these lasers are highly efficacious for dusting. Moreover, multiple fragments could directly be aspirated navigating the sheath without the need to popcorn. Additionally, we observe in our study that the lasing times and operative times are well within those that are previously reported in literature. ^24–25 Although we only report the laser settings and preferred lasing modalities and did not assess which setting is the best suited, given the promising results of newer technologies, it is important to indeed focus on laser parameters, parameter range, and the influence on ablation times and perhaps laser machine.

With regards to complications, Giulioni et al. showed from the FLEXOR study that ureteral injury, fever, and sepsis are still the most common complications. ²⁶ This is even more of a concern when using TFL or HPHL. However, in our study, the zero sepsis rate was achieved as FANS, by the integrated suction and aspiration, mitigates high intrarenal pressure ¹⁰ and temperature ¹¹ and reduces the risk of upper urinary tract injuries by providing good vision. In our study, surgeons reported that TFL was better for manipulation and visibility. The fibers used for TFL were 150–200 µm thick and 250 µm for HPHL. We argue that the property of the TFL fiber being more malleable, with less backburn and being thinner allowing for better irrigation as well as allowing more scope deflection could all be in some part contributing to this feature. ^27–28

The limitations of our study include a patient population without anatomical anomalies, the lack of assessment of the direct impact on costs and patient quality of life, lacking data on longer-term follow-up, and lacking stone analysis. In addition, there was no HL machine with MOSES technology employed and the technique used among the surgeons was not standardized. Finally, no data on efficiency parameters, such as ablation speed or ablation efficiency, were gathered. Despite these limitations, our results indicate that the use of FANS is safe and efficacious, with acceptable morbidity and no serious adverse events using both lasers. Notably, the complication and reintervention rates were significantly lower than those reported in previous large studies using conventional UAS. ²⁶

Conclusion

Our study shows that both lasers are safe, are efficacious, and can be suitably used for F-URS with FANS irrespective of stone location and burden. TFL shows a better ability to obtain a 100% SFR, and this may be relevant when tackling larger stone volumes. By combining with FANS, lasing and operative times were well within safe limits for both lasers making them equally good choices providing a high overall Grade A + B SFR.