A comparative study of fluoroscopy-guided closed reduction with 3D-printed guide-assisted cannulated screw fixation versus open surgery with a lag screw fixation in 32 cats with sacroiliac luxation–fracture

Abstract

Purpose

This prospective study aims to compare the clinical, radiographic, and computed tomographic (CT) findings of minimally invasive closed fluoroscopy-asisted reduction and cannulated screw fixation (MCF) versus open reduction (OR) with lag screws in cats with sacroiliac luxation (SIL).

Methods

Thirty-two cats with SIL were divided into two groups: MCF (n = 16) and OR (n = 16). In the MCF group, reduction was performed via a minimally invasive (MI) approach using fluoroscopy and a custom-made 3D K-wire guide, with stabilisation achieved using a 2.7 mm partially threaded cannulated screw. In the OR group, reduction was performed via an open approach, and stabilisation was achieved using a lag screw. Clinical, radiographic, and CT evaluations were performed for all cases.

Results

The MCF group demonstrated significantly shorter surgical duration (p < 0.001) and an earlier return to limb function compared to the OR group (p < 0.001). Lameness scores were significantly lower in the MCF group during the first two postoperative weeks (p < 0.001). Radiographically, the MCF group achieved significantly better restoration of the pelvic canal diameter ratio (PCDR) (Mean Difference = 0.091, 95% CI: 0.021–0.161, p = 0.012) and sacroiliac joint reduction ratio (SRO) (Mean Difference = 12.0, 95% CI: 5.76–18.20, p < 0.001). Although radiation exposure was higher in the MCF group (p < 0.05), the complication rate was significantly lower (18.8%) compared to the OR group (62.5%) (p = 0.029). The dorsoventral screw angle (DVA) (p = 0.012) and craniocaudal screw angle (CCA) (p = 0.632) were lower in the MCF group than in the OR group.

Conclusion

According to the results of this study the MCF technique using a guide is associated with shorter surgery duration, better anatomic reduction, fewer post-operative complications and faster recovery times. However, it does not seem to be a viable technique for delayed SIL cases in cats.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12917-026-05497-6.

Introduction

Sacroiliac luxation (SIL) in cats occurs frequently after trauma. Its incidence rate among pelvic fractures ranges from 59% to 93%. Previous studies have reported that bilateral SIL accounted for 27%–46% of cases, while unilateral SIL accounted for 55% [1–3]. SIL may be accompanied by sacral fractures [4]. Pelvic fractures may accompany unilateral SIL cases owing to the rigid box structure of the pelvis [5].

Conservative management is recommended if the cat is ambulatory, has no severe displacement of the articular surface, has minimal pain and has no associated fracture in the weight-bearing axis of the pelvic limb [6, 7]. However, as SIL occurs on the weight-bearing axis of the pelvic limb, some authors have suggested surgical management [7, 8]. Many techniques have been reported in the surgical management of SIL in cats, including the use of a lag screw, headless cannulated screw, tension band technique, external fixation and single transiliac pin or screw and nut [9–14]. SIL is traditionally stabilized with a lag screw [15–17]. Accurate positioning of the screw in the body of the sacrum is essential to prevent complications [18]. Postoperative complications, such as nerve damage, pain and screw loss, after incorrect screw placement have been reported [18, 19].

Two recent studies examined the minimally invasive (MI) treatment of SIL in cats and dogs using headless cannulated screws with intraoperative imaging guidance [10, 20]. In the veterinary literature, only one study with a few feline cases (4 cats and 26 dogs) compared the MI closed and open surgical approaches for the management of SIL [21]. However, no study has compared clinical, radiographic and computed tomography (CT) findings between the MI closed surgical approach with cannulated screw and the open surgical approach for the management of SIL in cats. Thus, aim of this study was to compare the clinical, radiographic and CT findings between open reduction (OR) and MI fluoroscopy-assisted closed cannulated screw placement in the surgical management of SIL in cats.

Materials and methods

This prospective comparative study was approved by the Local Ethics Committee for Animal Experiments of Ondokuz Mayıs University (HADYEK Number: 2023/84). The study was conducted and reported in accordance with the ARRIVE guidelines 2.0 (Animal Research: Reporting of In Vivo Experiments) to ensure transparent and complete reporting. Thirty-two cats with bilateral or unilateral SIL that were brought to the Veterinary Teaching Hospital of Ondokuz Mayıs University with bilateral or unilateral SIL between 2024 and 2025 were included in the study. Minimal invasive surgery was indicated for each cat; however, if closed reduction was not possible, open reduction was performed. Accoridng to the procedure performed, cats were allocated into two groups: the MI closed fixation (MCF) group (n = 16) and the OR group (n = 16). The patients' signalment (breed, age, sex and body weight), trauma aetiology, time elapsed after trauma and accompanying lesions were recorded. All the cats were followed up clinically and radiographically for at least eight weeks in the postoperative period.

Preoperative assessment

Right and left lateral, ventrodorsal and frog-legged pelvic radiographs were taken before surgery. Spiral CT (Mx16 evo CT scanner, Philips, Holland) scans were taken before surgery and immediately after surgery. The CT images were acquired at 120 kV and 240 mAs, with a 1-mm thickness. The CT analyses of the cases were performed on bone window scans (level 200 Hounsfield units, with 2.500 HU).

Anaesthesia and analgesia procedure

Dexmedetomidine (0.01 mg/kg IM) and butorphanol (0.2 mg/kg IM) were administered for premedication. After 10 min of premedication, an intravenous (IV) catheter was placed, and cefazolin (25 mg/kg IV) was administered. Propofol (4 mg/kg IV) was used to the induce of anaesthesia, which was maintained with isoflurane (2%) in an oxygen circular system (Mindray, Wato EX-20 Vet). Lidocaine 2% (4 mg/kg IT) was administered for epidural analgesia. A constant-rate infusion (12 mg ketamine and 48 mg butorphanol in 100 ml saline) was administered to all cats intraoperatively (5 mg/kg/h) and postoperatively (2 mg/kg/h) for analgesia. Atipamezole (0.2 mg/kg, IM) was administered to reverse the sedative effects of medetomidine and facilitate recovery from anaesthesia.

Custom-made three-dimensional k-wire guide

A three-dimensional (3D) custom-made Kirschner (K) wire guide was designed to facilitate the accurate placement of the guide pin into the sacrum. Modelling was performed using the Solid Works software (Dassault Systèmes, France), and the guide was printed with a polylactic acid filament (Creality K1 Max, Shenzhen Creality 3D Technology, Shenzhen, China). (Fig. 1) The printed 3D guide was customised with cerclage wires and K-wires to ensure that the entry and exit points could be clearly visualized under fluoroscopy and to prevent misalignment (Fig. 1B, C and D).

Fig. 1.

Illustrative and fluoroscopic views of the custom-made 3D-printed K-wire guide. (A) Craniolateral views of the guide. (B) Upper and (C) lateral views of the tip of the guide. Four grooves (green arrow) were created on the upper surface of the guide, into which four K-wires (green lines) were inserted to serve as radiopaque reference markers. In addition, two circular cerclage wires were fixed around the guide, one on the upper surface (blue circle and arrow) and another near the exit hole (red line and arrow), to ensure the entrance and exit points could be clearly visualised under fluoroscopy and to prevent misalignment. (D) An intraoperative fluoroscopic view demonstrates the guide placed over the sacral body, with the two cerclage wires (blue arrow: upper wire, red arrow: lower wire, and K-wire: green arrow) clearly visible for precise positioning

Surgical approach

All surgeries were performed by the same surgeon. The patient was positioned in sternal recumbency after the pelvis and both hindlimbs were prepared for aseptic surgery (Fig. 2A and B). Fluoroscopy-assisted closed reduction was performed using pointed bone-holding forceps by grasping the ischial tuberosity (Fig. 2C and D). The K-wire guide was placed in the sacrum, with the assistance of a custom-made 3D guide (Fig. 2G and H). Alternatively, if acetabular or ischial fractures were present on the affected side, the greater trochanter and iliac crest were grasped with pointed bone forceps to reduce the sacroiliac joint (Fig. 3). All surgical procedures in the MCF group are shown in Fig. 2. Open reduction and internal fixation using a lag screw were performed using the dorsal SIL approach in cases in which closed reduction could not be achieved [22]. All screws were placed in a lag manner in the OR group. For sacroiliac stabilization, 2.7-mm partially threaded cannulated screws (Aysam, Samsun, Turkey) and 2.4 and 2.7-mm fully threaded cortical screws with washers (Travmavet, Ankara, Turkey) were used. The surgical instruments included a washer, 2.2-mm cannulated drill bit, 0.8-mm K-wire and a cannulated screwdriver (Aysam, Samsun, Turkey) (Fig. 4). The surgery time for MCF was measured as the time from the placement of the first bone-holding forceps to the last suture. The operative time for OR was recorded as the time between the first incision and the last suture. All the patients were hospitalized for 24 h after the surgery.

Fig. 2.

All stages of the surgical procedure in Case 16 of the MCF group. Sternal recumbency position of the patient prior to reduction, with (A) intraoperative and (B) fluoroscopic views (OEC One C-arm, General Electric Healthcare, USA). (C-D) The ischial tuberosity was grasped on the affected side with pointed bone-holding forceps (red arrow) for traction of the ilium in the craniocaudal direction. Once optimal closed reduction was achieved, the reduction was maintained by holding the reduction forceps in place. (E) The incision site over the sacral body was marked with the no. 11 scalpel on the lateral ilium skin under (F) fluoroscopic visualization, and a stab incision was made extending from the skin to the iliac bone. (G) Placement of the 3D K-wire guide on the lateral ilium from the incision line and (H) lateral fluoroscopic view of the insertion of the 0.8 mm guide K-wire into the sacral body. (I) Drilling of the sacral body and ilium with a 2.2 mm cannulated drill bit by guidance of the K-wire and (J) fluoroscopic monitoring of the drill bit advancement within the sacral body. (K) Insertion of a 2.7 mm cannulated screw with a washer, and (L) fluoroscopic view of its advancement in the sacral body. (M) Ventrodorsal and (N) lateral fluoroscopic views after placement of the cannulated screw and removal of the guide wire. (O) Lateral and (P) ventrodorsal fluoroscopic views of the case after placement of the second screw using the same procedure technique

Fig. 3.

Fluoroscopic (A, B, D–F) and intraoperative (C) images of Case 10 with caudal acetabular fracture accompanied by sacroiliac luxation. (A) Before and (B) after reduction. (C) Pointed bone-holding forceps placed on the greater trochanter (red arrow) and iliac crest (blue arrow) for sacroiliac reduction. Lateral fluoroscopic views of (D) the 3D custom-made guide positioned on the lateral sacral body and (E) the 0.8 mm K-wire placed in the sacral body. (F) Final postoperative fluoroscopic view of the sacroiliac joint

Fig. 4.

The view of surgical instruments (a) 0.8-mm K-wire, (b) 2.2-mm cannulated drill bit, (c) washer, (d) 2.7-mm partially threaded cannulated screw, (e) 2.4-mm cortical screw, (f) universal drill guide, and (g) cannulated screwdriver

Postoperative care

Meloxicam (0.05 mg/kg PO SID) was administered for 4 days for postoperative pain management. Amoxicillin–clavulanic acid (25 mg/kg PO BID) was administered for 10 days.

Postoperative management consisted of a graded activity restriction protocol. Owners were instructed to enforce strict cage confinement for the first 2 weeks after surgery to ensure initial soft tissue healing and implant stability. This was followed by an additional 2 weeks of small-room confinement with strict restriction from jumping and running. This 4-week controlled activity period aligns with conservative management recommendations for pelvic fractures [6, 23]. Clinical and radiographical evaluations were performed on the first and seventh postoperative days and every 2 weeks for at least the 8 weeks after surgery. A visual analysis was used for lameness evaluation of all cases. (1 = no lameness, 2 = intermittent lameness, 3 = frequent weight-bearing lameness, 4 = persistent weight-bearing lameness and 5 = non-weight-bearing lameness) [24].

Postoperative radiographic evaluations

Sacral screw purchase (SSP) rate, sacroiliac joint reduction ratio (SRO), pelvic canal diameter ratio (PCDR) and hemipelvic canal width ratio (HPCW) were calculated from the ventrodorsal pelvic radiographs taken preoperatively and immediately after surgery. All pelvic measurements were performed as described in a previous study [10]. The optimal SSP rate was ≥ 60% [25]. The SRO was calculated as a percentage and was considered optimal at ≥ 90% [18]. The PCDR was considered normal at ≥ 1.1 [18]. The HPCW was measured as the ratio of the side of the sacroiliac dislocation to the unaffected side, and pelvic symmetry was reported as 1 [18]. All postoperative radiographic measurements are presented in Supplementary Fig. 1.

Postoperative computed tomographic evaluations

Spiral CT scans were obtained immediately after surgery. Multiplane reconstruction was used in the CT measurements. The angle between the screw and the dorsal plane was measured as the DVA. In addition, the angle between the transversal plane and the screw was measured as the CCA [17]. In the DVA measurements made in the transversal plane, a positive angle indicates the dorsal plane, while a negative angle indicates ventral screw deviation. In CCA measurements made in the dorsal plane, positive and negative angles indicate cranial and caudal screw deviations, respectively. The dorsoventral articular angle (DAA) was measured in the transverse plane on the CT images by measuring the angle between the sacroiliac joint surface and the screw within the sacral body [26]. Radiographic measurements were performed using the DICOM software (Horos, Nimble Co., Version 4.0.1) (Supplementary Fig. 2).

Statistical analysis

The normality of the data distribution was assessed using the Shapiro–Wilk test. A two-way repeated-measures analysis of variance was used to evaluate the interaction between the reduction method (closed vs. open) and radiographic measurements (PCDR, HPCW and SRO) obtained before and after surgery, with Bonferroni-adjusted pairwise comparisons. Independent samples t-tests were performed to compare the trauma-to-surgery interval, SSP rate, DVA, CCA, DAA, fluoroscopy time (seconds), number of fluoroscopic views, radiation dose (mGy) and dose–area product (DAP, Gy·cm²) between the groups. The Mann–Whitney U test was used for comparisons of surgery duration and time to first limb use. The changes in lameness scores across the postoperative time points were analysed using the Friedman test, followed by Wilcoxon signed-rank tests for pairwise comparisons. The between-group differences in lameness scores at each time point were assessed using the Mann–Whitney U test. In the cats with bilateral SIL, the time to first limb use was also compared between the techniques using the Mann–Whitney U test because of the non-normality of the data distribution. The complication rates between the MCF and OR groups were analysed using the Fisher’s exact test. Statistical analyses were performed using SPSS Statistics (Version 21.0, IBM Corp., Armonk, NY, USA).

Results

A total of 40 sacroiliac fractures and luxations (fractures: 4, luxations: 36) in 32 cats (closed: 16, open: 16) were included in this study. The mean ages, mean body weights and pelvic fracture types of all the cases are summarised in Supplementary Table 1. Domestic shorthair (93.75%) was the most frequent breed among the cats included in the study. The most common cases of injury were road accidents (16/32) and falling from a height (12/32). All the patients included in the study had an American Society of Anaesthesiologists physical status classification of grade I.

Intraoperative findings

The surgery duration and the time between the onset of trauma and surgery (TBTS) using the reduction technique are summarised in Table 1. The surgery duration was shorter in the MCF group than in the OR group (p < 0.001). The TBTS was categorized into two groups: 0–3 days and 4 days. The surgery durations were 24.37 ± 14.33 min (0–3 days) and 24.37 ± 14.33 min (≥4 days) (p = 0.990) in the MCF group and 35.83 ± 9.85 min (0–3 days) and 45.55 ± 21.36 min (≥4 days) in the OR group (p = 0.178).

Table 1.

Comparison of time between the onset of trauma and surgery, surgery duration, and time to first use of the limb between groups. Data are presented as mean ± SD or median (min-max)

Pre-intra-and postoperative measurements	MCF Group	OR Group	P-value
Time between the onset of trauma and surgery (days)	2.87 ± 1.54	3.3 ± 1.49	0.422
Surgery duration (minutes)	24.42 ± 12.27	34.00 (25–90)	< 0.001*
Time to first use of the limb (days)	1.00 (1–3)	2 (1–3)	< 0.001*

MCF Closed surgical technique, OR Open surgical technique

* P < 0.05 indicates statistical significance

The cannulated screw placement procedure began with the insertion of the guide wire, and a hole was created in the sacrum body using a cannulated drill. During this stage, if the drill bit passed beyond the guidewire, the guidewire was pulled back along with the drill bit, while the drill bit was removed from the sacrum body. In such a situation, drilling was stopped to prevent the guide pin from coming out and to maintain the reduction. Subsequently, the cannulated drill bit was detached from the drill chuck, and the guidewire was advanced further (Fig. 2J and L). Afterwards, the cannulated drill bit inside the sacral body was removed, and the procedure was continued.

The K-wire guide was bent in the sacrum in Case 12 during the drilling stage because of the slight angulation of the drill bit while the K-wire guide was inside. At this point, as the tip of the drill bit was at a level ≥60% of sacrum within the sacral body, the drill bit was removed, and the guide wire was left in the sacral body. A cannulated screw was applied under the guidance of the guidewire. During the removal of the K-wire guide, the bent K-wire broke off at the level of the screw head and was left in the sacral body (Fig. 5A and D). The patient was followed up for 60 days, but no complications occured.

Fig. 5.

Clinical and radiographic views of Case 12 (A–D), Case 11 (E–I), and Case 16 (J–P). Intraoperative fluoroscopic views of Case 12, (A) the view of the K-wire before drilling. (B) The drilling stage of the sacrum, and (C) after the view of the bent K-wire. (D) Fluoroscopic view of the bent K-wire that broke off. (E) Clinical view of Case 11 showing unilateral proprioceptive deficit in the right hindlimb. (F) Preoperative and (G) postoperative radiographic views of Case 11. (H) Transverse and (I) dorsal plane CT scans immediately after surgery. (J) Immediately postoperative ventrodorsal radiographic view of Case 16 with bilateral sacroiliac luxation. (K) Postoperative transverse CT image showing screw penetration into the sacral canal (red arrow). (L) Intraoperative fluoroscopic view at postoperative day 6 demonstrating implant failure. (M) Ventrodorsal and radiographic view after revision surgery with transiliac and transsacral pinning. (N) Ventrodorsal radiographic view on postoperative day 19 after removal of the transiliac pin because of failure of the pin. (O) Ventrodorsal radiographic and (P) clinical views on postoperative day 170

The mean fluoroscopy times were 112.98 ± 42.98 s and 44.57 ± 20.92 s in the MCF and OR groups (p < 0.001), respectively. In both groups, the bilateral cases generally had a greater fluoroscopy time than the unilateral cases. The mean number of fluoroscopy images taken and the cumulative radiation exposure levels were higher in the MCF group than in the OR group (p < 0.05) (Table 2). A mean of 2.6 times more intraoperative images were required for optimal screw placement in the MCF group (p < 0.001). All personnel wore personal dosimeters throughout the surgical procedure. No readings were found in the dosimeter results.

Table 2.

Comparison of fluoroscopy parameters (Mean ± SD) including image number, exposure time, and cumulative radiation levels in the MCF and OR groups

Number of fluoroscopic image (n)	MCF Group (n = 16)	OR Group (n = 16)	P-value
	79.5 ± 38.94	29.75 ± 15.90	< 0.001*
Fluoroscopy time (sec)	156.79 ± 84.38	71.40 ± 30.34	< 0.001*
Examination dose (mGy)	5.26 ± 3.26	2.47 ± 1.12	0.003*
Examination DAP (Gy cm²)	1.83 ± 2.40	0.46 ± 0.25	0.031*

MCF Closed surgical technique, OR Open surgical technique

* P < 0.05 indicates statistical significance (Independent t-test)

Postoperative findings

The cats in the MCF group (1.10 ± 0.45 days) began using their legs earlier after surgery than those in the OR group (1.85 ± 0.85 days) (p < 0.001) (Table 1). On the first day after surgery, 93.75% of the cats in the MCF group and 50% of those in the OR group began using their limbs. In cats with bilateral SIL, the mean initial weight-bearing time for both legs was 1 day in the MCF group and 2.5 days in the OR group, which indicates a significant delay (p = 0.041). The median lameness scores on the first, seventh and fourteenth day after surgery were significantly lower in the MCF group than in the OR group (p < 0.001). The lameness score linearly decreased after surgery in both groups (p < 0.05). On the twenty-first day after surgery, the difference in the lameness scores between the groups was not statistically significant (p > 0.05). The median follow-up times were 61 days (range, 35–526 days) and 64.50 days (range, 56–300 days) in the MCF and OR groups, respectively (Table 3).

Table 3.

Comparison of postoperative lameness scores (median [min–max]) between cats treated with Closed and Open surgical techniques, analyzed using the Mann–Whitney U test

Days	MCF Group Median (min–max)	OR Group Median (min–max)	P-value
Preoperative	4.0 (4–5)	4.0 (4–5)	0.670
1st day	3.0 (2–5)	5.0 (3–5)	< 0.001*
7th day	1.0 (1–4)	3.0 (2–4)	< 0.001*
14th day	1.0 (1–2)	2.0 (1–3)	0.003*
21st day	1.0 (1–2)	1.0 (1–3)	0.101

MCF Closed surgical technique, OR Open surgical technique

* P < 0.05 indicates statistical significance

Radiographic image analysis

The PCDR increased postoperatively in both groups, but the increase was statistically significant only in the MCF group (p < 0.001). Furthermore, the postoperative increase in PCDR was significantly greater in the MCF group than in the OR group (Mean Difference = 0.091, 95% CI: 0.021–0.161, p = 0.012) (Fig. 6A). The HPCW did not show a statistically significant increase postoperatively (p = 0.067). The difference in the change in HPCW from before to after surgery between the MCF and OR groups was not significant (p = 0.254) (Fig. 6B). Furthermore, pairwise comparisons revealed no significant difference between the MCF and OR groups (Mean Difference=-0.020, 95% CI:-0.129–0.088, p = 0.706). The SRO significantly increased postoperatively in both groups (p < 0.05). Pairwise comparisons showed that the increase was significantly greater in the MCF group than in the OR group (Mean Difference ≈ 12.0, 95% CI: 5.76–18.20, p < 0.001) (Fig. 6C). These findings indicate that closed reduction is more effective than open reduction in restoring sacroiliac alignment and enlarging the PCDR, with the HPCWs remained largely unaffected. The SSP rate was calculated to be in the MCF group and 70.52% in the OR group, (p = 0.085).

Fig. 6.

Comparison of preoperative and postoperative radiographic measurements in cats managed with closed and open surgical techniques. (A) Pelvic Canal Diameter Ratio, (B) Hemipelvic Canal Width Ratio and (C) Sacroiliac Joint Reduction Ratio. Data are presented as mean ± SD. * indicates statistically significant difference (p < 0.05); ns indicates non-significant (p > 0.05)

Computed tomographic protocol and image analysis

The mean DVA was calculated as − 2.17° ± 5.35 in the MCF group and − 8.24° ± 8.85 in the OR group (p = 0.012). (Supplementary Fig. 2) The mean CCA was significantly lower in the MCF group (− 4.41° ± 4.46) than in the OR group (− 5.33° ± 7.25) (p = 0.632). The MCF group demonstrated a higher mean DAA (98.17°) compared with the OR group (93.40°) (p = 0.171). Computed tomographic assessment verified that all screws were accurately positioned without breaching the sacral canal.

Complications

Complications were observed in 13 (40.6%) of the 32 cats (Table 4). The incidence was significantly lower in the MCF group (3/13) than in the OR group (10/13) (p = 0.029). Most complications (9/13) were resolved without the need for additional surgical intervention, including proprioceptive deficits (8/9) and urinary incontinence (1/9, Case 32) (Fig. 5E and I), three of the six cats in the OR group that developed proprioceptive deficits (Cases 21, 26 and 29) had bilateral SIL. Proprioceptive deficits typically improved within 2 weeks after surgery. Complications that required surgical management (5/13) included implant failure (1/5) and suture dehiscence with wound infection (4/5). In Case 16, concurrence between, suture dehiscence and implant failure was observed (Fig. 5J and P).

Table 4.

Distribution of complications according to groups and corresponding case numbers

Complication	MCF Group (n = 16)	OR Group (n = 16)	P-value
Proprioceptive deficit	2/16 (12.5%)	6/16 (37.5%)	0.002
Urinary incontinence	0/16 (0%)	1/16 (6.25%)	1.000
Implant failure	1/16 (6.25%)	0/16 (0%)	1.000
Suture dehiscence and superficial wound infection	1/16 (6.25%)	3/16 (18.75%)	0.599
Total	3/16 (18.75%)	10/16 (62.5%)	0.029*

MCF Closed surgical technique, OR Open surgical technique

* P < 0.05 indicates statistical significance (Fisher’s Exact Test)

Bilateral SIL in eight cases was managed with MCF (Cases 9, 11 and 16) and OR (Cases 21, 22, 24, 26 and 29). Postoperative CT evaluation confirmed that the screws were entirely contained within the safe zone of the sacral body without violation of the sacral canal or neuroforamen in all cases, with the exception of Case 16. In Case 16, a postoperative CT scan revealed the presence of a right screw in the transverse plane, within the sacral canal (Fig. 5K). Immediately after the CT scan, the cat underwent cardiopulmonary arrest; thus, the procedure for screw realignment was suspended. On the sixth postoperative day, implant failure occurred in the screw of the sacral canal, and the screw was removed under anaesthesia (Fig. 5L). After removing the failure screw, the SIL was managed using transilial and transsacral pin methods to avoid damaging the sacral body (Fig. 5M). On the nineteenth postoperative day, the transilial pin was removed because of suture dehiscence and transilial pin implant failure (Fig. 5N). The patient was observed for a period for 170 days and made a full recovery (Fig. 5O and P).

Discussion

The cats that underwent closed reduction had a shorter surgery time, more favourable outcomes, lower infection rates and relatively decreased lameness scores than the cats that underwent open reduction. Conversely, the TBTS emerged as a critical determinant of the feasibility of the closed technique. We observed that closed reduction became significantly more challenging if the TBTS exceeded 4 days. This difficulty is primarily attributed to the progressive contraction of the pelvic musculature and the organization of the hematoma into fibrous tissue, which resists non-invasive manipulation. Therefore, from a clinical decision-making perspective, the MCF technique should be prioritized for acute cases. In patients presenting with a delay of more than 4 days, surgeons should be aware of the decreased likelihood of achieving successful closed reduction and prepared to convert intraoperatively to OR to ensure anatomical alignment.

Fluoroscopy-guided closed reduction and sacroiliac stabilisation have been reported in the veterinary literature as viable methods [21, 27, 28]. Initial studies in this area have primarily focused on cadavers, defining safe corridors and appropriate screw application techniques [17, 29–31]. Many studies have been conducted in both cats and dogs [10, 15, 20, 28]. Open versus closed reduction of SIL was previously studied in a small population, and favourable clinical outcomes were observed. Detailed patient data, such as lameness score and radiographic evaluation, on a large number of cats are lacking in veterinary literature [21]. The larger sample size, perioperative and postoperative findings, lameness scores and radiographic and CT findings suggest that closed reduction is indeed favourable but not always possible in cats, as mentioned in the previous study.

The reduction and stabilization of SIL using MI techniques have been reported to provide more optimal screw placement and lower screw loosening rate and a lower complication rates [21, 31]. In a previous study, the mean surgical duration of closed reduction was 90 min [28]. In our study, the duration was significantly shorter. If the reduction was deemed too difficult, we immediately switched to the open technique, which could be the reason for the decrease in surgery duration. The use of a 3D K-wire guide definitely expedites the process. Moreover, this difference from previous studies may be explained by the fact that not using cannulated screws might have prolonged the surgical duration. Furthermore, in a cadaver study, the reduction of the experimentally created SIL was performed in a shorter time [30]. The TBTS prolonged the surgical duration in the OR group, and a similar surgical duration was measured in the MCF group.

In ex vivo studies, the usefulness of a 3D drill guide for stabilizing SIL using an MI technique was compared with with that of free hand, OR and internal fixation techniques [19, 32, 33]. In a cadaver study that compared fluoroscopy-assisted iliosacral screw placement in dogs using freehand and 3D drill guide applications, the use of 3D guidance eased the procedure [34]. According to a previous study, stated that the table-mounted reduction system and 3D-printed drill guide MI technique stabilisation in cats with SIL enables optimal screw placement [35]. However, owing to the retrospective nature of the previous study, a control group was lacking. The use of the guide increased accuracy and decreased the screw placement duration by a factor of 1.34 times compared with open reduction, reducing the overall time of the entire procedure, including anaesthesia. The area on the cat’s sacrum where the screw can be correctly placed has been defined as < 0.5 cm² [14]. Thus, the correct screw placement in cats has been reported in 33%–87.5% of cases [18, 25]. As there is a relatively narrow safe implant corridor in the sacrum of cats [14], it can be difficult even in open surgery to directly visualize the sacroiliac joint surface during surgery, drill and place the lag screw without a guide. Furthermore, the open surgical technique requires an invasive approach. The 3D K-wire guide used in this study could be applied to the narrow sacral body using fluoroscopy. In intraoperative lateral pelvic fluoroscopic imaging, the sacral canal and L7-S1 disk space were also identified prior to guidewire placement. Fluoroscopic guidance reduced the postoperative screw loosening rate and surgery duration in the stabilization of SIL in cats using an open surgical approach [27]. However, previous cadaveric studies have reported a shorter fluoroscopy time than that in our clinical study [30, 36], which could be attributed to the effort required to reduce luxation in a live patient.

In a cadaveric study, the freehand and 3D drill guide numbers of the fluoroscopy images were similar [34]. In our study, freehand was used only in the OR group; thus, the fluoroscopy time was significantly shorter than that in closed reduction, mainly because fluoroscopy is not as essential when directly viewing the anatomical landmarks. A critical concern regarding fluoroscopy-guided minimally invasive procedures is the potential exposure of the surgical team to radiation. Although the MCF group required significantly longer fluoroscopy time and higher fluoroscopy dose than the open technique group, no measurable radiation exposure was detected by the personal dosimeters worn by the surgical team throughout the study. This finding confirms that adherence to standard radiation safety protocols effectively mitigates occupational risks, thereby establishing the safety profile of the MCF technique for operating personnel. In addition, although the sample size precluded a statistical sub-analysis, an observational trend of decreased fluoroscopy use was noted in the later cases of the series. This ‘learning curve’ effect suggests that as the surgeon’s familiarity with the 3D guide and imaging workflow increases, the efficiency of the procedure improves. In spite of the greater need for intraoperative radiology in the MCF group, the surgery took less time, and screw placement accuracy and SRO were higher in this group.

In our study, the cats in the MCF group began using their limbs earlier and had a lower lameness score than those in the OR group. No significant difference in the lameness score was found between the MCF and OR groups in the third week after surgery. This short-term lameness may be attributed to the more invasive nature of the OR group and the tissue damage resulting from surgery.

In previous studies, SIL was evaluated using radiological measurements of the PCDR, HPCW, SSP rate and SRO [21, 26, 28]. Previous studies have reported wide variability in SSP rate for both open and closed techniques in SIL repair [21, 26, 27, 30, 31]. In our study, complications were more common in the OR group, with screw loosening occurring in one case (3.1%). Unlike earlier reports that showed high incidence rates of screw loosening in open procedures, this study achieved an SSP rate > 60% all the cats, which likely contributed to the low complication rate. The SSP rates was significantly higher in the MCF group, likely due to the use of fluoroscopy in open surgery and the 3D custom-made K-wire guide in closed procedures, both of which supported accurate screw placement. Regarding the biomechanical properties of the implants, a potential concern with cannulated screws is their smaller core diameter than that of the standard cortical screws used in the OR group, which might suggest lower bending strength. However, biomechanical stability is not solely determined by implant dimensions but also by the quality of bone purchase. In the MCF group, the 3D-printed guide ensured precise placement of the screw into the ‘safe zone’ of the sacral body, which contains the densest bone stock. This positional accuracy maximizes screw pull out strength and effectively compensates for the reduced core diameter of the cannulated screw, ensuring robust fixation stability comparable with that of standard lag screws.

The sacroiliac reduction ratio in ex vivo studies using a closed MI surgical approach ranges from 87.73% to 98.33%, on average [26, 30]. In vivo studies have reported sacroiliac reduction ratios ranging from a mean of 83.2% in a study that used an open surgical approach to a mean of 71.8%–86.6% in a study using an MI surgical approach [27, 28]. Another in vivo study reported that sacroiliac reduction ratios > 90% were achieved in 54.16% of cases that underwent an open surgical approach and 38.46% of cases that underwent a closed approach [21]. Our SRO measurements were better than those in previous vivo studies [21, 27, 28] but were lower than those in one ex vivo study [26].

Ex vivo and in vivo studies have reported that an optimal PCDR is achieved after the stabilization of SIL [21, 26, 28, 30]. The PCDR measurements in ex vivo studies are generally higher than those in vivo studies, and this discrepancy has been attributed to the technique being more challenging to apply in clinical cases because of the presence of pelvic and extremity fractures [21]. As the cats with failed closed reduction were reassigned to the OR group, the remaining MCF cases could inherently have more favourable PCDR potential. Moreover, the K-wire guide used in MCF helped maintain reduction until screw placement, thereby improving the PCDR outcomes.

DVA, CCA and DDA measurements in the CT evaluation of SIL have been reported in ex vivo [26, 31], and in vivo [28], studies. In one ex vivo study, the mean DVA and CCA in dogs that underwent open and MI surgical approaches were significantly lower in the MI surgery group [31]. In a cadaveric study, the mean DDA ranged from 96.24° to 98.65° [26], similar that the value in our study. In a study involving five cats with SIL treated with MI surgical techniques, CT measurements showed a mean DVA of 1.3° and a mean CCA of − 4.5° [28]. In our study, the DVA and CCA were lower in the cases treated with closed reduction.

This study has several limitations. First, the study design was not a strict randomized controlled trial. Cases in which closed reduction could not be achieved were shifted to OR. While clinically necessary, this conversion might have introduced selection bias, as the open reduction group likely included more complex or delayed cases. Moreover, concomitant orthopaedic injuries in some cats might have influenced the clinical lameness scores. The variation in the TBTS also presented a challenge. Specifically, delayed admission increased the difficulty of closed reduction due to soft tissue contraction. Furthermore, although the high reduction quality (SRO) observed in this study suggests a favourable long-term prognosis, long-term (6–12 months) radiographic follow-up was not available for all cases. This limitation prevents the definitive evaluation of late-stage complications, such as the progression of osteoarthritis or implant fatigue. Lastly, a prior power analysis was not conducted in this study because of the prospective clinical nature and specific caseload constraints. However, significant differences in the primary outcome measures (e.g. fluoroscopy time and joint reduction rates) were observed, indicating that the sample size was sufficient to detect these effects.

This study shows that in cats, when closed reduction of SIL in cats was possible, the application of an MI surgical technique with a percutaneously placed cannulated screw resulted in a shorter surgery duration an that of the open surgical technique and was a more efficient method efficient method in terms of using the affected limb. Postoperative care was easier in the cats that underwent closed surgical techniques, and the postoperative complication rate was low. Optimal screw placement could be achieved with fluoroscopy support and a 3D custom-made K-wire guide, and cats that underwent this technique returned to their daily routine earlier after surgery. However, it is important to note that the implementation of this technique requires specific infrastructure, including intraoperative fluoroscopy and 3D printing capabilities, which may limit its widespread accessibility in general veterinary practice. Nevertheless, in clinics equipped with these resources, placing a percutaneous cannulated screw using a 3D custom-made K-wire guide is a superior method to open surgical techniques. The closed surgical technique should be preferred in cats, including those with bilateral SIL, because of its favourable clinical outcomes it provides.

Abbreviations

MCF

Minimally Invasive Closed Fluoroscopy-assisted Reduction and Cannulated Screw Fixation

OR

Open Reduction

SIL

Sacroiliac Luxation

CT

Computed Tomography

3D

Three-Dimensional

MI

Minimally Invasive

TBTS

Time Between The Onset of Trauma and Surgery

SD

Standard Deviation

PCDR

Pelvic Canal Diameter Ratio

SRO

Sacroiliac Joint Reduction Ratio

HPCW

Hemipelvic Canal Width Ratio

SSP

Sacral Screw Purchase

DAP

Dose–Area Product

DVA

Dorsoventral Screw Angle

CCA

Craniocaudal Screw Angle

DAA

Dorsoventral Articular Angle

HU

Hounsfield Units

DICOM

Digital Imaging and Communications in Medicine

IV

Intravenous

IM

Intramuscular

PO

Per Os

SID

Semel in Die

BID

Bis in Die

Authors’ contributions

Conception of study: K.S.İ., and C.N. Study design and acquisition of data: K.S.İ., C.N., E.B.K. and T.Ö. Data analysis and interpretation: K.S.İ., and C.N.

Funding

This study was supported by the Ondokuz Mayıs University Scientific Research Projects Commission [Grant numbers: BAP06-2024-4880] The funders had no role in the study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Data availability

The datasets generated and/or analysed during the current study are available from the corresponding author on reasonable request.

Declarations

Ethics approval and consent to participate

This prospective study was approved by the Local Ethics Committee for Animal Experiments of Ondokuz Mayıs University (HADYEK Number: 2023/84). The informed consent was obtained from all patient owners.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.