Pelvic osteotomy for pelvic canal stenosis after malunion pelvic fractures in cats

Abstract

Objectives

The aim of this study was to assess the efficacy of pelvic osteotomy and ventral fixation of the ischium using cortical screws and polymethylmethacrylate (PMMA) for feline pelvic canal stenosis (PCS) associated with malunion after conservative management of pelvic fractures.

Methods

Surgical pelvic enlargement was performed for PCS in six cats. The medical records, including information on the patients, surgical procedures, defecation and complications, were reviewed. The sacral index (SI) and colonic:lumbar:vertebral ratio (CLVR) were evaluated based on pre- and postoperative radiographs.

Results

This study included five castrated male cats and one spayed female cat. Postoperative improvements in constipation and defecatory difficulty were noted in all cases. The postoperative SI was significantly higher (mean 0.93, range 0.72–1.13) than the preoperative SI (mean 0.59, range 0.45–0.74) (P <0.001). However, no statistically significant change was found in the CLVR preoperatively and up to 3 months postoperatively. A successful union of the ilium was observed, without implant failures. One case developed necrosis of the pubic surgical wound.

Conclusions and relevance

This study indicated the potential benefits of pelvic osteotomy and ventral fixation of the pelvic floor using screws and PMMA for achieving pelvic cavity enlargement in treating feline PCS associated with defecatory problems.

Introduction

Orthopaedic pelvic injuries in cats reportedly include iliac body, pelvic floor, acetabular and ischial fractures, as well as sacroiliac fractures/luxation.^1,2 Typical pelvic injury combinations in cats include the following: iliac body fracture with contralateral sacroiliac joint dislocation and pelvic floor fracture; iliac body and pelvic floor fractures with unilateral sacroiliac joint dislocation; or bilateral sacroiliac joint dislocation and pelvic floor fracture. Such multiple orthopaedic pelvic injuries considerably destabilise the pelvic structure.^1,2 If left untreated, these fracture combinations can lead to the collapse of the ventral (ischium and pubis) and caudal (beyond the iliac body) components of the pelvis, causing malunion and pelvic canal stenosis (PCS).^3,4

PCS resulting from malunion after pelvic fracture contributes to 25% of secondary megacolon cases.^4–6 Research suggests that surgical intervention to enlarge the pelvic cavity is recommended if constipation due to PCS is present for less than 6 months.^7,8 If constipation persists for more than 6 months, subtotal colectomy is recommended.^4,7–9 Although various surgical approaches for PCS have been reported, few reports are available on the short- to long-term follow-up evaluation.^3,10–13

The aim of this study was to introduce a novel surgical technique for pelvic enlargement, combining corrective osteotomy of the ilium and pubis with ventral fixation of the pelvic floor using screws and polymethylmethacrylate (PMMA), in cats with PCS and to evaluate its clinical efficacy based on short- to long-term postoperative outcomes. We hypothesised that this method would avoid postoperative restenosis of the pelvis and improve defecatory function.

Materials and methods

Cases

Cases of defecatory difficulty and PCS after pelvic fractures managed at the Veterinary Medical Teaching Hospital of Nippon Veterinary and Life Science University between 2010 and 2019 were included in this study.

Details collected from the medical records included age, weight, sex, defecation status, duration of constipation, surgical method and postoperative complications. Pelvic radiography was performed pre- and postoperatively (see radiographs of cases 1, 2 and 4–6 in the supplementary material), and in some cases, CT was performed preoperatively. Defecatory function, based on owner interviews, was categorised into the following three grades: grade 1 (natural defecation without assistance); grade 2 (no manual removal of faeces but requiring lactulose as oral stool softener); and grade 3 (intermittent or continuous manual removal of faeces necessary). Complications were classified as major if they necessitated additional surgical intervention and minor if not.

Surgery

Anaesthesia was induced using propofol (7 mg/kg IV) and maintained with inhalational isoflurane and oxygen after endotracheal intubation. Fully controlled respiratory management employed a pressure-controlled ventilator (maximum airway pressure 8–15 cm H₂O, inspiratory: expiratory ratio 1:2–1:4). Perioperative measures included lactated Ringer’s solution infusion (3–5 ml/kg/h) for blood volume maintenance, epidural administration of 0.5% bupivacaine hydrochloride (0.2 ml/kg) and morphine hydrochloride (0.1 µg/kg), and continuous fentanyl infusion (1–3 µg/kg/h) for analgesia. Antibiotics included cefmetazole (25 mg/kg IV), administered approximately 1 h before skin incision and every 90 mins intraoperatively. ¹⁴ All surgeries were performed by one surgeon (YH). After induction of general anaesthesia, the caudal lumbar area and both dorsal and ventral aspects of the pelvis were shaved and aseptically prepared. The cat was positioned in dorsal recumbency and a ventral approach to the pelvis was made. A sagittal saw (Colibri II; DePuy Synthes) was used to create an osteotomy of the pelvic floor. Subsequently, the patient was placed in lateral recumbency to enable surgical exposure of the lateral aspect of the ilium, which was severely displaced into the pelvic cavity. ¹⁵ After identifying the caudal edge of the sacroiliac joint, an osteotomy was made perpendicular to the long axis of the ilium.

The caudal fragment of the ilium was retracted craniolaterally; a reconstruction (MIZUHO) or custom-made crank plate (Platon Japan Co) was used to stabilise the osteotomy site (Figure 1). Crank plates were available with four, seven or eight screw holes and allowed the use of 2.4 mm cortical screws. The crank plate available with seven screw holes had two and one holes on the cranial and caudal sides, respectively, with locking features. All crank plates were designed to achieve 7.5 mm lateral displacement of the caudal fragment (Figure 2). After wound closure, the patient was placed in dorsal recumbency again and the separation of the ventral pelvic osteotomy was confirmed. Anchor screws were inserted into the ventral aspect of the pubis and ischium. To protect the pelvic cavity organs, such as the ureters and rectum, absorbent haemostatic material (AVITENE; Davol) and a pedunculated fat graft, harvested subcutaneously, were placed into the osteotomy site. Liquid PMMA bone cement (Surgical Simplex; Stryker Japan KK) was applied to incorporate the anchor screws and form a ventral fixation support. During bone cement polymerisation, lavage with 4°C physiological saline was performed to avoid thermal polymerisation damage. The wound was closed once the bone cement had fully solidified.

Figure 1.

(a–d) Radiographic and (e–h) intraoperative findings of case 3. Preoperative (a) ventrodorsal and (b) lateral aspects of the pelvis are shown. Postoperative (c) ventrodorsal and (d) lateral aspects of the pelvis are shown. In the ventrodorsal view, the left ilial body is displaced medially from the caudal edge of the sacroiliac joint; narrowing of the pelvic cavity is clearly observed. The width of the sacrum at the cranial border (SW) and across the width of the pelvic canal at the narrowest point, between the medial cortices of the acetabulae (AW), are indicated by arrows. These measurements are used to calculate the sacral index (SI), which is the ratio of these two measurements. Lateral radiographs show enlargement of the descending colon with marked faecal impaction and deformed fusion of the ilial body. Postoperative radiography reveals pelvic cavity enlargement. The maximum diameter of the colon (CD) and the length of L5 (L) are delineated by solid black bars. (e) Yellow dotted lines indicate the separation line of the pelvic floor osteotomy; (f) the ilium is osteotomised and fixed with a crank plate (seven holes); (g) eight anchor screws are inserted into the ventral aspect of the pelvic floor; (h) placement of polymethylmethacrylate

Figure 2.

A design drawing of a typical crank plate. This shows an eight-hole crank plate. All crank plates are designed to facilitate a 7.5 mm lateral displacement of the caudal bone fragment after ilial osteotomy. Measurements on the drawing are in mm

A fentanyl patch (12.5 μg/h, Duprop MT patch; Janssen Pharmaceutical) was applied to provide analgesia immediately postoperatively. Fentanyl was continuously infused until the patch was effective and was tapered as necessary. The patch was removed 3 days postoperatively. Postoperative management included a 2-week cage rest period and administration of cefmetazole 25 mg/kg IV q12h.

Radiographic evaluation

Bone union after osteotomy of the ilium was assessed using postoperative radiography. Bone union at the osteotomy site of the pelvic floor could not be evaluated because of the presence of PMMA. The integrity of the plates and screws was evaluated at the ilial osteotomy site. It seems unlikely that PMMA was destroyed at the pelvic floor; therefore, the integrity of the screws used was evaluated. Radiographs were taken immediately postoperatively, and at 1.5, 3, 6, 12 and 24 months postoperatively and every year thereafter for cases in which long-term follow-up was possible.

Sacral index

The pelvic shape was assessed using ventrodorsal radiography (Figure 1). The sacral index (SI) was calculated as the ratio of the narrowest width of the pelvic canal (AW) to the width of the cranial sacrum (SW), measured between the medial cortices of the acetabulae (SI = AW/SW). ³

Degree of pelvic canal narrowing

Pelvic canal narrowing was quantified following the methodology described by Hamilton et al. ³ The SI was calculated as the width of the pelvic canal divided by that of the cranial sacrum. The SI of normal cats has been reported to be 0.97. ³ Therefore, the estimated width of the pelvic canal before injury can be calculated by multiplying the normal SI by the cranial sacral width of each case. ³ The degree of pelvic canal narrowing is expressed as the ratio of the original width of the pelvic canal before injury to the width of the pelvic canal after injury. However, since the original width of the pelvic canal before injury cannot be measured, the calculated estimated width of the pelvic canal before injury was used. Subsequently, the degree of pelvic canal narrowing was expressed as a percentage, using the estimated width of the pelvic canal before injury and the narrowed width of the pelvic canal after injury for each case.

Colonic:lumbar:vertebral ratio

The colonic:lumbar:vertebral ratio (CLVR), an index used for assessing defecatory abnormalities, ⁶ was assessed based on an abdominal lateral radiograph encompassing the lumbar vertebrae and pelvis (Figure 1). The CLVR was calculated as the ratio of the maximum colon diameter to the length of the L5 vertebral body. The colon’s maximum diameter was measured from the ilial tuberosity to the pelvic entrance at its most distended point. ⁶

Statistical analysis

The SI and CLVR were measured by one examiner (RS). They were also measured by radiographic images using OsiriX-Lite software version 14.0.1 (Pixmeo). Pre- and postoperative SIs and CLVRs were analysed using repeated measures analysis of variance with mixed methods in SAS software version 9.3 (SAS 2011; SAS Institute). The Tukey–Kramer test adjusted for multiple comparisons was used, with significance set at P <0.05. Analyses were conducted preoperatively, immediately postoperatively and 2 weeks, 1.5 months and 3 months postoperatively, acknowledging that regular long-term follow-ups were not available for some cases.

Results

Case information and outcomes

Six cats (five castrated males and one spayed female), weighing 3.6–5.5 kg (mean 4.3 kg), were included in this study. Detailed information of each case is presented in Table 1. Five cats were confirmed to have pelvic malunion upon rescue and subsequently developed PCS. One cat, initially treated conservatively for a pelvic fracture upon rescue, subsequently developed PCS and defecatory problems. All cases had confirmed pelvic malunion and PCS. However, their fracture timelines and post-injury clinical histories were unclear. Of the six cats, five underwent pelvic CT. Details of the orthopaedic injuries are summarised in Table 2. The plates used were a reconstruction plate in one case, a four-hole crank plate in one case, a seven-hole crank plate in one case and an eight-hole crank plate in three cases (Table 2). Preoperatively, all cases were classified as having grade 3 defecatory function. Three months postoperatively, defecatory function improved to grades 1 and 2 in five and one cases, respectively. Two cases of persistent defecatory problems for 2–3 years showed postoperative improvements. The only major complication involved necrosis of the subcutaneous tissue of the pubic surgical wound (case 4).

Table 1.

Characterisation and evaluation of six cats with pelvic canal stenosis

Case	Breed	Sex	Age (years)	Body weight (kg)	Grade of defecatory abnormality	Duration of clinical signs	SI		Pelvic canal narrowing (%)		Colonic lumbar vertebral ratio		Maximum follow-up period
							Preoperative	Postoperative 3 months	Preoperative	Postoperative 3 months	Preoperative	Postoperative 3 months
1	Mixed	SF	4	3.7	3	3 years	0.45	0.72	53.3	26.1	1.51	1.12	3 months
2	Mixed	CM	2	3.6	3	5 months	0.43	1.13	55.2	−16.8	1.07	1.03	3 years
3	Mixed	CM	Estimated 4	3.6	3	2–3 years	0.61	1.04	37.5	−6.9	1.43	1.18	3 months
4	Mixed	CM	Age unknown	5.4	3	Unknown	0.76	0.90	21.6	6.9	0.96	1.06	1.5 years
5	Mixed	CM	Estimated 5	3.8	3	Unknown	0.66	1.05	32.4	−8.3	1.16	0.91	4.5 months
6	Mixed	CM	Estimated 2	5.5	3	Unknown	0.50	0.71	48.7	27.2	1.27	0.90	4 years

The normal value of the sacral index (SI) is 0.97. SI = width of pelvic canal/width of cranial border of sacrum

CM = castrated male; SF = spayed female

Table 2.

Details of the orthopaedic injuries and implants used in the six cases

Case	Fracture site	Details of implants
		Body of the ilium	Pelvic floor
1	Left sacroiliac luxation, right femoral neck fracture, left acetabular fracture, fracture of the pubis and ischium	2.0 mm reconstruction plate (MI) ∅2.0 mm cortical screw ×6 (J&J)	∅1.5 mm cortical screw ×2 (J&J) ∅2.0 mm cortical screw ×6 (J&J) PMMA
2	Left ilial body fracture, fracture of the pubis and ischium	Titanium crank plate (4 × 4 hole) (PJ) ∅2.4 mm titanium cortical screw ×4 (PJ)	∅1.5 mm cortical screw ×10 (J&J) PMMA
3	Left ilial body fracture, fracture of the pubis and ischium	Titanium crank plate (4 × 3 hole) (PJ) ∅2.4 mm titanium cortical screw ×7 (PJ) ∅2.4 mm titanium locking screw ×7 (PJ)	∅1.5 mm cortical screw ×8 (J&J) PMMA
4	Right sacroiliac luxation, fracture of the pubis and ischium	Titanium crank plate (2 × 2 hole) (PJ) ∅2.4 mm titanium cortical screw ×4 (PJ)	∅1.5 mm cortical screw ×6 (J&J) PMMA
5	Left ilial body fracture, fracture of the pubis and ischium	Titanium crank plate (4 × 4 hole) (PJ) ∅2.4 mm titanium cortical screw ×4 (PJ)	∅1.5 mm cortical screw ×8 (J&J) PMMA
6	Right sacroiliac luxation, fracture of the pubis and ischium	Titanium crank plate (4 × 4 hole) (PJ) ∅2.4 mm titanium cortical screw ×4 (PJ)	∅1.5 mm cortical screw ×8 (J&J) PMMA

J&J = Johnson & Johnson; MI = MIZUHO; PJ = Platon Japan Co; PMMA = polymethylmethacrylate

∅ = screw diameter

Radiographic assessment

Five of the six cases (cases 1–3, 5 and 6) exhibited healing of the iliac osteotomy site within 2–6 months postoperatively. In case 4, bone union was confirmed 9 months postoperatively. Throughout the observation period, no failure of metal implants or PMMA was noted.

SI

The evaluation of pre- and postoperative radiographs revealed no pelvic restenosis (Table 1). The mean preoperative SI was 0.57 (range 0.43–0.76), whereas the mean postoperative SI was 0.93 (range 0.72–1.13), indicating a significant postoperative increase (P <0.001) (Table 3). In three cases (cases 2, 4 and 6) with long-term follow-up (1.5–4 years postoperatively), no pelvic restenosis was observed.

Table 3.

Comparison of preoperative and postoperative SI and CLVR in the six cases

	Preoperatively	Immediately postoperatively	2 weeks postoperatively	1.5 months postoperatively	3 months postoperatively
SI	0.57 ± 0.08a	0.90 ± 0.08b	0.92 ± 0.08b	0.94 ± 0.08b	0.93 ± 0.08b
CLVR	1.23 ± 0.06a	1.02 ± 0.06a	1.18 ± 0.06a	1.01 ± 0.07a	1.03 ± 0.06a

Same lowercase letters indicate no significant difference between columns (significance level: P <0.05)

Data are mean ± SE.

CLVR = colonic:lumbar:vertebral ratio; SI = sacral index

Degree of pelvic canal narrowing

Before the corrective surgery, the mean pelvic canal narrowing was 41.5% (range 21.6–55.2). Case 4 exhibited moderate narrowing (10–30%) and five cases (cases 1–3, 5 and 6) showed severe narrowing (>30%) (Table 1). Three cases (cases 1, 2 and 6) had a narrowing of ⩾45%, categorised as high risk for constipation-related issues. ¹³ Postoperatively, the mean degree of narrowing was reduced to 4.7% (range −6.9 to 27.2), and pelvic cavity enlargement was confirmed in all cases.

CLVR

The mean preoperative CLVR was 1.23 (range 0.96–1.51). The CLVRs of four cases (cases 2 and 4–6) were below 1.28: one (case 3) was in the range of 1.28–1.48 and another (case 1) exceeded 1.48 (Table 1). The mean postoperative CLVR decreased to 1.03 (range 0.90–1.18). None of the cases showed a CLVR >1.48 (indicating a megacolon) after surgery. ⁶ No significant difference was found in the maximum colon diameter to L5 length ratio from preoperatively to 3 months postoperatively (Table 3).

Discussion

This study evaluated a surgical procedure for achieving pelvic enlargement in cats with PCS, combining pelvic osteotomies and internal fixation of both the ilial and midline osteotomy sites. No case of restenosis of the pelvic cavity was observed postoperatively, and clinical signs improved even in cases with long-term defecatory problems.

Corrective surgical methods for malunion associated with PCS have been reported, including the use of plates, ⁸ external skeletal fixators, ¹⁶ lag screws ¹³ and spacers.^17,18 However, most case reports have lacked long-term follow-up assessments. Schrader ⁸ evaluated long-term clinical signs after pelvic enlargement but did not include an anatomical evaluation of the pelvic cavity. To our knowledge, no previous reports of combined corrective osteotomy of the ilium and midline pelvic floor with ventral fixation of the pelvic floor using screws and PMMA are available. The present study provides a short- and long-term assessment of both the clinical and anatomical features of pelvic enlargement surgery in cats. PMMA, which is characterised by high compressive strength compared with tensile strength, ¹⁹ counteracts forces that may cause medial displacement of the caudal ilial osteotomy fragment. It was hypothesised that by combining ventral fixation of the pubis and ischium, the force that pulls the caudal ilial fragment into the pelvic cavity at the ilial osteotomy site could be neutralised. This study demonstrated a significant postoperative improvement in the SI compared with preoperative values in all cases, without significant pelvic restenosis observed up to 3 months postoperatively. In addition, in three cases of long-term follow-up (approximately 1.5–4 years postoperatively), SI measurements over time indicated no signs of pelvic cavity restenosis compared with immediate postoperative values. Furthermore, no instance of plate fracture or screw loosening was observed throughout the observation period in any case. In 2009, Hamilton et al ³ showed that in approximately half of the cases treated with ilial plating alone, screw loosening occurred within 4–6 weeks, followed by postoperative pelvic stenosis.

A crank or reconstruction plate was used to enlarge the narrowed pelvic cavity in the lateral direction while displacing the osteotomised caudal fragment outward. To prevent restenosis of the pelvic cavity, anchor screws were placed on the ventral aspect of the pelvic floor (ischium and pubis) and these were fixed with PMMA. Notably, two cases (cases 1 and 3), which had constipation and defecatory problems for 2–3 years, showed improvement of clinical signs postoperatively. Despite needing lactulose as an oral stool softener postoperatively, case 1 exhibited improved clinical signs. No significant difference was observed between pre- and postoperative CLVRs. A CLVR >1.48 is indicative of megacolon. ⁶ In case 1, this ratio was 1.51 preoperatively but did not exceed 1.48 postoperatively, suggesting that pelvic cavity enlargement could alleviate long-term constipation and defecation difficulties. Moreover, surgical treatment could facilitate the management of the constipated patient by periodic manual evacuation. In small animal clinical practice, surgical pelvic cavity enlargement is recommended for PCS-induced constipation lasting for less than 6 months,^7,8 whereas subtotal colectomy is advised for durations exceeding 6 months.^4,7–9 Our results highlight the potential benefits of pelvic cavity enlargement and reconstruction in resolving PCS, even in cases where constipation secondary to PCS has persisted beyond 6 months.

In this study, a crank plate was used in five cases and a reconstruction plate in one case at the ilial osteotomy site. The crank plate allows a 7.5 mm lateral shift of the caudal fragment relative to the cranial fragment, facilitating reliable pelvic enlargement. However, as the size of the ilium, location of the initial fracture and severity of malunion will vary in individual cases, preparing specific crank plates for each case might be necessary. Currently, crank plates are custom-made to order and may not be immediately available to the surgeon. Therefore, it may require some time to obtain them. Care should be taken when using an alternative plate in place of the crank plate, as excessive bending may reduce rigidity.

A major postoperative complication, subcutaneous tissue necrosis of the ventral surgical wound, was observed in case 4. However, this was effectively managed with surgical debridement. Roh et al ¹⁸ reported urinary dysfunction as a potential complication after symphyseal distraction osteotomy using PMMA for PCS. This was attributed to tension in the ischiocavernosus muscle (originating from the caudal surface of the ischial tuberosity and inserting into the bulbospongiosus muscle) ²⁰ and excessive pelvic enlargement. In our study, no case of postoperative urinary dysfunction or peripheral neuropathy was noted. This may be because our technique involves unilateral osteotomy of the ilium with lateral translation of the caudal fragment while simultaneously separating the pubic symphysis. As a limitation of the study, the number of cases was small and the follow-up period was short. Further studies with a larger number of cases and longer follow-up assessments are warranted.

Conclusions

This study describes a pelvic enlargement technique in cats for PCS. This approach combined osteotomies of the ilium and midline pelvic floor and used crank (or reconstruction) plates to stabilise the ilial osteotomy and screws and PMMA to stabilise the pelvic floor osteotomy.