Femoral head and neck ostectomy via ventral approach in cats using a standardized three-dimensional-printed ostectomy guide

Abstract

Objectives

This study aimed to develop a standardized ostectomy guide for ventral femoral head and neck ostectomy (vFHO) in cats. We aimed to assess the guide’s accuracy for maximizing removal of a clinically acceptable amount of bone without sacrificing soft tissue attachments.

Methods

CT scans of 18 cats with normal femoral morphology were obtained to determine ideal ostectomy planes and generate three-dimensional (3D)-printed femurs. A standardized ostectomy guide was designed, printed and used to perform the ostectomies on printed bones as well as on three cadavers via vFHO. Postoperative CT scans were obtained. Covariates including age, sex, neuter status, body weight, side and neck inclination, and version angles were recorded. Ostectomies were assessed by comparing actual vs ideal ostectomy angles and percentage of femoral head and neck removed using CT scan data, and by blinded evaluation of printed bones with vFHO. Mixed-effects models were used for statistical analysis.

Results

The ideal and actual ostectomy angles in the 3D-printed femur models were statistically equivalent (P <0.005), whereas the percentage of femur removed was not (P = 0.080) until two outlier specimens (smallest/youngest) were excluded, after which both measurements became statistically equivalent (P = 0.001). No covariates significantly influenced outcomes. Visual assessment by a blinded surgeon found 75% of ostectomies to be acceptable; unacceptable ostectomies had either over-removal of the greater trochanter or incomplete neck resection. Cadaveric testing confirmed the guide’s usability, with practical application and adequate exposure via a previously described ventral approach.

Conclusions and relevance

This study demonstrated that although a standardized vFHO ostectomy guide in cats was practical to use and produced ostectomies statistically equivalent to ideal cuts, occasional over- or under-resection and the risk of damaging the greater trochanter preclude clinical application at this stage, highlighting the need for further design optimization.

Plain language summary

Using a three-dimensional-printed guide for hip surgery in cats

This study looked at whether a specially designed three-dimensional (3D)-printed guide could help veterinary surgeons perform a type of hip surgery in cats more accurately. The surgery involves removing part of the hip joint – the femoral head and neck – using a ventral (underside) approach. The goal was to make precise cuts without harming important soft tissues.

To test this, researchers used CT scans from 18 healthy cats to figure out the ideal way to make these cuts. They then created 3D-printed models of the cats’ thigh bones and designed a guide to assist with the procedure. The guide was tested on these printed bones and on three cat cadavers. After the surgeries, CT scans were used to compare the actual cuts with the ideal ones. They also checked whether factors like age, weight or bone shape affected the results.

The guide generally helped achieve cuts close to the ideal ones, especially after removing two very small cats from the analysis. Most surgeries were judged acceptable by an independent reviewer, but some had issues – either too much bone was removed or not enough. Tests on cadavers showed the guide was practical to use with the ventral approach.

The guide shows promise for making hip surgeries in cats more accurate, but the design needs more work before it can be used in clinics.

Introduction

The femoral head and neck ostectomy (FHO) is a common salvage procedure performed for a multitude of pathologies of the coxofemoral joint in cats.^{1 –6} The FHO is indicated for conditions including chronic coxofemoral osteoarthritis, fractures, luxation, avascular necrosis and failed total hip replacement.^1,2,6,7 During FHO, the femoral head and entire neck are excised to (1) limit bony contact between the proximal femur and acetabulum, (2) alleviate pain associated with movement of a diseased coxofemoral joint and (3) establish a fibrous pseudoarthrosis.^1,2,6 Suboptimal outcomes may occur from bone-on-bone contact between the medial aspect of the proximal femur and the dorsal acetabular rim, leading to pain and lameness.^{1,3,4,6,8 –10} Successful postoperative FHO outcomes rely on achieving an ideal balance of maximizing bone removal without sacrificing soft tissue attachments.

An FHO can be performed via a craniolateral (clFHO) or a ventral (vFHO) approach.^4,8 The clFHO is more commonly used, in part because of surgeon comfort, as the clFHO is utilized for other procedures and avoids major neurovascular structures. The craniolateral approach does require dissection of the dorsal support structures required for establishment of a pseudoarthrosis, including incision of the craniodorsal joint capsule and, if additional exposure is necessary, a partial tenotomy of the deep gluteal muscle. ¹¹ Disruption of these structures may result in reduced support of the joint and, potentially, an inferior functional outcome for the patient.^4,12 Additional dissection required for this approach includes elevation of the quadriceps from the proximal femur, allowing visualization of both major bony landmarks for FHO, the greater and lesser trochanters. Alternatively, the benefits of the vFHO include (1) sparing the dorsal soft tissue structures of the hip, (2) better visualization of the lesser trochanter without elevation of the quadriceps femoris and (3) elimination of the need for repositioning when bilateral FHOs are performed.^4,8 Contraindications for performing vFHO may include craniodorsal luxation, ventral or caudal pelvic pathology, and cases needing dorsal acetabular exposure. In addition to the restricted visualization of the greater trochanter during vFHO, the surgeon must also be cognizant of the presence of the femoral artery and vein, the medial circumflex femoral artery and vein, as well as the femoral/saphenous nerve within the surgical approach, and protect these structures. Ultimately, the development of a surgical cutting guide that is practical to implement, promotes accurate and consistent vFHO, and is applicable for use by both general practitioners and board-certified surgeons may improve the surgical efficacy of this less common approach.

We established two objectives: (1) to develop a three-dimensional (3D)-printed, standardized ostectomy guide for vFHO in cats based on a CT database; and (2) to assess the accuracy of the FHO using the guide on 3D-printed femurs. We hypothesized that (1) a standardized ostectomy guide could be designed using our feline femoral morphology database, which (2) would generate an ostectomy statistically and clinically equivalent to the ideal ostectomy in 3D-printed feline femurs.

Materials and methods

Case selection

A radiology database was used to search for feline cases between 2010 and 2024 (Carestream Vue Motion Archive Explorer, Version 12.2.1.600164; Carestream Health). Of the 96 feline CT scans reviewed, 18 met the inclusion criteria, defined as studies with clear visualization of the femoral head, neck, both trochanters, acetabulum and proximal femoral shaft. Studies with an incomplete femur (n = 8) were used for guide design, and studies with a complete femur (n = 10) were printed and used for guide application. Additional information recorded included case ID, signalment, weight, date of birth and CT date. In addition, three feline cadavers, euthanized for reasons unrelated to the study, were utilized to ensure the guide could be applied via the ventral surgical approach.

Pre- and post-ostectomy CT on printed models and cadavers

CT was performed on printed models after the ostectomy, and for the cadavers before and after the ostectomy, using an LB CT (Aquilion LB; Toshiba). CT scans were performed at 120 kVp and 150 mAs with a scanned coverage of 320.00 mm. The displayed field of view was 32.0 × 43.1 cm. Multiplanar reconstructions of the femur were analyzed using image viewing software (Version 12.2.1.600164; Carestream Health). The femoral head and neck were defined as the region proximal to a line connecting a point just proximal to the lesser trochanter and the proximomedial aspect of the greater trochanter at the intertrochanteric notch.

Guide design

The 3D-printed ostectomy guide was created using a cloud-based computer-aided design software platform (OnShape, 2024; Version 1.186.42034.db603590db80) based on the average femoral shape of cases (n = 8) that were not used for performing ostectomies (Table 1). Right and left versions of the guide were printed using polylactic acid filament (PLA; BambuLab) on a Bambu Lab P1P printer (BambuLab) (Figure 1).

Table 1.

Measurements of the cases used for guide development

Case number	Signalment at time of CT	Weight at time of CT (kg)	Length of left femur– anatomic axis (mm)	Length of right femur – anatomic axis (mm)	Width of left femur at level of proximal lesser trochanter (mm)	Width of right femur at level of proximal lesser trochanter (mm)	Width of left femur at level of distal lesser trochanter (mm)	Width of right femur at level of distal lesser trochanter (mm)	Ideal left femoral head and neck ostectomy angle	Ideal right femoral head and neck ostectomy angle
1	6 yo FS DSH	4.30	75.13	74.57	11.95	11.61	10.37	10.30	28.16	25.35
2	6 yo MC DSH	2.78	88.96	89.20	11.52	11.49	9.27	9.17	24.41	24.24
3	7 yo MC Bengal	5.60	96.23	87.81	13.81	14.12	12.97	12.96	28.57	29.51
4	5 yo FS DLH	3.68	77.21	79.36	11.81	11.44	9.87	10.16	25.74	24.14
5	11 yo FS DSH	4.15	104.95	106.75	13.26	12.87	11.33	11.33	31.14	27.95
6	2 yo FS DSH	5.90	71.23	72.10	12.20	11.72	10.49	10.72	21.48	18.64
7	9 yo FS DSH	5.16	12.28	12.42	15.38	16.27	13.43	13.43	29.45	29.78
8	10 yo MC Manx	4.84	11.26	11.11	14.25	14.51	11.70	13.16	27.78	27.67
Mean		4.55	67.16	66.67	13.02	13.00	11.18	11.40	27.09	25.91
Median		4.57	76.17	76.96	12.73	12.29	10.91	11.02	27.97	26.51
Range		2.78–5.90	11.26–104.95	11.11–106.75	11.52–15.38	11.44–16.27	9.27–13.43	9.17–13.43	24.41–31.14	18.64–29.78

DLH = domestic longhair; DSH = domestic shorthair; FS = female spayed; MC = male castrated; yo = years old

Figure 1.

(a) Digital guide design. The base lies along the medial cortex of the femur, secured by 0.045 mm Kirschner wires (K-wires), and the flange interacts with the lesser trochanter. An arch cutout allows the guide to be placed over the iliopsoas. Based on preliminary measurements of feline femurs, the flange was angled at 28° off the guide base. (b) Virtual application of the right osteotomy guide on a right femur. X = lesser trochanter

Femoral bone models

Bilateral femurs for cases with complete femurs included on CT scans were printed (10 cases, 20 femurs). Segmentation was performed using Materialise Mimics (Version 26.0.0.576; Materialise), and resultant stereolithographies were printed on a FormLabs 3BL Medical printer (Formlabs; 3BL Medical [PKG-F3BL-WO-MSP-BASIC]) using Photopolymer Resin Durable (FLDUCL02; FormLabs) with post-processing per the manufacturer’s guidelines.

Ostectomy of femoral bone models

Printed models were secured in a vice with the medial surface upright. The guide’s lesser trochanter ledge was placed distally, with the flange flush against the proximal lesser trochanter. A surgical drill (VetSurgicalSolutions Mini Driver Multifunction Surgical Drill, NM3-6001E; BlueSAO) placed a 1.6 mm × 230 mm Kirschner wire (K-wire) (Trocar Smooth 316 LVM Stainless Steel; IMEX) into the proximal and distal holes of the guide, through the cis cortex and seated at the trans cortex. The K-wires were cut approximately 1 cm above the guide. The ostectomy was performed using the surgical drill with a sagittal saw attachment; the saw blade was laid flush against the guide flange.

Ostectomy of the cadaveric femurs

Successful placement of the guide in the cadaveric model was defined as follows: (1) the guide could be positioned and secured to the femur using K-wires through an approach that simulated vFHO in a live patient, including that no musculature needed to be transected; and (2) the flange offered a sufficient surface area for performing the ostectomy without compromising adjacent soft tissues. The cadaver was placed in dorsal recumbency and a medial surgical approach to the proximal left femur was performed as previously described (Figure 2), without transection of the pectineus or any other musculature.^4,11 The femoral head was exposed by incising the ventral joint capsule and transecting the round ligament with a #15 blade (Veterinary Supply Co, MWI). The iliopsoas was gently elevated to allow placement of the guide’s ledge against the proximal lesser trochanter, with the iliopsoas positioned beneath the guide’s arch. The guide was secured with proximal and distal 1.6 mm × 230 mm K-wires. A sagittal saw was laid flush against the flange to perform the ostectomy, and the femoral head and neck were excised.

Figure 2.

Cadaveric ventral surgical approach to the left femur (a) without and (b) with the three-dimensional-printed ostectomy guide in place; cranial is to the right. # = femoral head; * = iliopsoas muscle; X = lesser trochanter; f = femur; P = pectineus muscle; A = adductor muscle

CT analysis

The ideal and actual area removed were measured using the multiplanar reconstruction of each sample’s CT scan in the Carestream PACS software. The transverse plane was positioned on the lesser trochanter and adjusted to be perpendicular to the femur’s long axis. The sagittal plane was rotated to become an oblique plane that bisected both the greater and lesser trochanters, while remaining perpendicular to the transverse plane. The ‘closed curve region of interest’ (in mm²) feature was selected to measure the entire visible area (femoral head, neck, greater trochanter, lesser trochanter and any visible metaphysis and diaphysis), which was recorded as the ‘total area of femur’. Next, the ‘total area of ideal femur removed’ was measured as all bone proximal to a line connecting the proximal aspect of the lesser trochanter and the proximomedial aspect of the greater trochanter (Figure 3). The ideal and actual bone removal from the cross-sectional area of bone removed, as a percentage, was calculated.

Figure 3.

(a) Total area of femur. (b) Ideal area of removal. (c) Actual area remaining. AR = area; AV = average; SD = standard deviation

Ideal and actual FHO angle

The ideal FHO angle (iFHOA) and actual FHO angle (aFHOA) were measured and recorded as previously described ⁸ using the volume reconstruction in Carestream PACS software. Briefly, the long axis of the femur was defined and, on a craniocaudal view, a second line was made crossing the long axis of the femur from just proximal to the greater trochanter to just proximal to the lesser trochanter. The acute angle formed at the intersection of the two lines defined the iFHOA. Postoperatively, the digital models were aligned identically and the aFHOA was measured.

Neck inclination and version angles

To assess the femoral morphology of the sample population, neck inclination and version angles were measured preoperatively as previously described.^{13 –15}

Blinded ostectomy assessment

A third, blinded surgeon (CF) visually assessed the 3D-printed femur models after the ostectomy and subjectively determined if the ostectomy was clinically acceptable. The criteria used to determine if a cut was acceptable included adequate removal of the femoral neck and minimal removal of the greater trochanter. If the cut was determined to be unacceptable, a brief explanation was provided.

Statistical analysis

To evaluate equivalence between actual vs ideal ostectomy angle and actual vs ideal bone removal percentage, we used two one-sided t-tests. Our equivalence bounds were set at –5° and +5° and –5% and +5% for the lower and upper bounds, respectively, as a representation of potentially clinically acceptable error range (Figure 4). In addition, we used a linear mixed-effects model to assess whether any covariates (neck inclination angle, neck version angle, body weight and age), factors (left or right femur) or cluster variables (animal) influenced the accuracy of the FHOs. Statistical significance was defined as P <0.05.

Figure 4.

Right femur showing ideal cut (black) and upper (+5°) (red) and lower (–5°) (blue) bounds representing the potentially clinically acceptable range

Results

The iFHOA, aFHOA, and the ideal and actual percentages of femur removed for the 3D-printed femur models are summarized in Table 2. The mean aFHOA (28.6° ± 5.2) was statistically equivalent (P <0.005) to the mean iFHOA (30.0° ± 4.5) (Figure 5). The mean actual percentage of femur removed (41.2% ± 9.6%) was not statistically equivalent to the ideal percentage (38.4% ± 4.4%) (P = 0.080) (Figure 6). Data review suggested the smallest/youngest specimens (cases 9 and 10) may be outliers, as they were more than 2 SDs from the mean. When the analysis was repeated excluding these four femurs, the aFHOA (28.6° ± 5.22) and iFHOA (30.0° ± 4.52) remained equivalent (P = 0.001), and the actual (41.2% ± 9.60%) and ideal (38.4% ± 4.44%) percentages of femur removal became equivalent (P = 0.001).

Table 2.

Covariates/factors and measured outcomes after guided simulated vFHO on printed bones

Case number	Sex	I vs N	Body weight (kg)	Age (years)	L vs R	Neck inclination angle	Neck version angle	iFHOA	aFHOA	Ideal % of femur removed	Actual % of femur removed	Blinded assessment
1	M	N	6.16	9.00	L	106.55	19.01	34.00	31.00	35.12	35.23	Acceptable
1					R	102.79	19.47	32.00	31.50	35.27	35.35	Acceptable
2	M	N	6.00	8.00	L	119.83	18.17	23.00	34.00	35.28	35.92	Unacceptable
2					R	117.12	16.42	25.50	25.00	40.45	40.45	Acceptable
3	M	N	4.70	6.00	L	97.86	19.18	35.50	32.00	34.36	36.73	Acceptable
3					R	96.16	27.34	32.50	35.50	30.77	29.61	Acceptable
4	F	N	4.60	6.00	L	98.43	25.47	30.00	26.00	34.13	37.20	Acceptable
4					R	94.53	25.50	32.00	28.00	34.72	40.23	Unacceptable
5	M	N	4.51	4.00	L	115.98	26.36	28.00	23.50	34.67	30.97	Acceptable
5					R	105.98	31.13	31.00	28.00	36.94	41.38	Acceptable
6	F	N	3.70	10.00	L	104.98	25.04	31.00	24.50	40.69	40.74	Acceptable
6					R	98.20	24.69	31.00	32.00	44.05	45.93	Acceptable
7	F	N	3.00	1.00	L	111.73	15.98	20.00	26.00	44.33	37.79	Unacceptable
7					R	105.75	12.98	20.50	18.00	37.45	35.09	Acceptable
8	F	N	2.97	8.00	L	112.24	24.62	35.50	27.50	39.41	39.32	Acceptable
8					R	115.82	22.15	32.00	31.00	35.58	35.34	Acceptable
9	F	I	1.50	0.25	L	108.75	24.67	35.00	29.00	40.19	40.93	Acceptable
9					R	113.96	21.68	30.00	30.00	42.33	63.04	Acceptable
10	F	I	0.91	0.15	L	106.95	30.36	31.00	19.50	47.18	61.91	Unacceptable
10					R	97.82	29.01	31.00	40.00	44.70	60.26	Unacceptable
11	F	N	4.87	15.00	L	120.19	14.04	32.00	29.00	39.90	44.02	–
12	M	N	2.50	7.00	L	101.30	23.60	32.00	31.00	38.60	38.05	–
13	F	N	2.56	13.00	L	110.75	18.11	31.00	25.00	41.00	41.27	–
	Mean					107.12	22.50	30.07	28.60	38.73	41.43	–
	Median					106.55	26.30	31.00	29.00	38.60	39.32	–
	Range					94.53–120.19	12.98–31.13	23.00–35.50	18.00–40.00	30.77–47.18	29.61–63.04	–

aFHOA = actual FHO angle; F = female; FHO = femoral head and neck ostectomy; I = intact; iFHOA = ideal FHO angle; M = male; N = neutered; vFHO = ventral FHO

Figure 5.

Distribution of mean differences of angles with confidence intervals. FHO = femoral head and neck ostectomy

Figure 6.

Distribution of mean differences of percentages removed with confidence intervals

No significant effects of covariates or factors (including side, body weight, age or femoral neck angle measurements) were found to influence the accuracy of the FHO angle (Table 3) or the percentage of bone removed (Table 3).

Table 3.

Results of mixed-effects models determining whether covariates/factors influenced the accuracy of the femoral head and neck ostectomies (FHOs)

Covariate/factor	aFHOA *	% bone removed †
Left vs right	0.222	0.194
Body weight (kg)	0.525	0.372
Age (years)	0.549	0.685
Neck inclination angle	0.795	0.918
Neck version angle	0.454	0.424

^*P value of linear mixed-effects model where the significance value was set to alpha = 0.05

^†P value of linear mixed-effects model where the significance value was set to alpha =  0.05

aFHOA = actual FHO angle

When the 3D-printed femurs were examined visually by a blinded participant (CF) after simulated vFHO, 15/20 (75%) were considered to be acceptable osteotomies (Table 2). Of the five femurs that were not considered acceptable, three had excess greater trochanter removed and two had too much femoral neck remaining (Figure 7). The two femurs with residual neck remaining were the right and left femurs of the youngest/smallest sample.

Figure 7.

(a) Excess femoral neck remaining. (b) Excess greater trochanter removed

The cadaveric procedures demonstrated that the guide could be applied to the proximal medial femur via the surgical approach, without any muscular transection, including that the entire flange was against the bone at the level of the proximal lesser trochanter.

Discussion

Achieving the ideal ostectomy plane is critical for optimizing clinical outcomes after FHO. We aimed to develop a standardized cutting guide for vFHO in cats to provide an accurate ostectomy through this less common approach while preserving dorsal musculature. We hypothesized that a standardized guide would produce an ostectomy equivalent to the planned angle and bone removal. We partially accepted our hypothesis: the ideal and actual angles were found to be equivalent, while the ideal percentage of bone removed was less than the actual amount removed. The guide was also easily applied to cadaveric specimens via the ventral approach, allowing K-wire fixation and ostectomy with minimal disruption to surrounding structures.

The iFHOA in our samples was 26° in the population of cats used for guide design (n = 8) and 30° in the population of cats used for testing the guide (n = 10). A retrospective morphological survey of 100 cats reported a mean ±SD iFHOA of 33.7° ± 4.3 (range 22.0–54.3). ¹⁶ This survey used radiographs to determine the angle and therefore the accuracy of measurement was inevitably affected by radiographic positioning. ¹⁶ Future guide designs should include a larger sample size for determination of flange angle.

The actual percentage of bone removed was 2.8% higher than the ideal percentage of bone removed, but the aFHOA was equivalent to the iFHOA. This discrepancy was mild and could be due to differences in our ability to accurately position the planes and models and measure these two outcomes. However, if this discrepancy is accurate, it suggests the guide produced the correct FHO angle but was positioned slightly distal to the ideal site, resulting in greater bone removal. Alternatively, it is possible that some of this discrepancy may be secondary to potential outliers, namely the two smallest and youngest specimens. When these samples were removed, the actual and ideal percentages of bone removal were equivalent.

An experienced blinded orthopedic surgeon evaluated the clinical acceptability of the ostectomies, and 15/20 (75%) ostectomies were deemed clinically acceptable. Out of five unacceptable cuts, three involved the removal of some of the greater trochanter, which could impair gluteal muscle or rotator attachments and have a negative clinical effect (Figure 8a–c). Out of five cuts, two were deemed unacceptable because of the presence of residual femoral neck (Figure 8d–f) that could cause persistent bone-on-bone contact and pain; from a ventral approach, it is likely that this could be identified intraoperatively and additional bone removed to address this finding. Both cases with residual femoral neck came from the youngest and smallest specimens, suggesting either a need for varying guide sizes or that extremes in femoral morphology, including size, may reduce the applicability of a standardized guide. Immature cats rarely require FHO, but this variation warrants further study. This outcome highlights the need for surgeon awareness even when using surgical guides, although it is also possible that this outcome could be mitigated with various guide refinements such as a slightly steeper flange angle or the use of a cutting slot rather than a flange to limit the potential for the surgeon to lift the saw blade off the flange, resulting in a flatter cut angle. Alternatively, the guide may have been positioned too far distally on the femoral shaft, leading to cuts that incorporated the greater trochanter, rather than the blade’s tilt being the cause of the unacceptable cuts.

Figure 8.

Representative samples of (a–c) too much greater trochanter removed from the right femur and a clinically acceptable cut of the left femur; and (d–f) not enough femoral neck removed near the lesser trochanter of the right and left femurs. (a,d) Cranial view. (b,e) Medial view. (c,f) Oblique, externally rotated view. L = left; R = right

One reason for pursuing vFHO is for improved visualization of the lesser trochanter with less dissection of the dorsal support tissues and quadriceps. Residual lameness after FHO has been speculated to be from incomplete resection of the medial femoral neck. One study found no correlation between incomplete resection and decreased ground force reaction (GFR), ¹⁷ while a second study found that a majority (12/17, 70%) of cats with incomplete neck resection had a lower GFR than other cats. ¹⁰ Retention of the lesser trochanter is purported to improve stability and maintain normal hip function; therefore, it is not recommended by the authors to remove the lesser trochanter unless necessary.

Ultimately, the decision between vFHO vs clFHO depends on surgeon preference, as either approach can achieve accurate ostectomy while avoiding muscular transection, requiring just muscle separation and joint capsulotomy. To date, one cadaveric study has assessed the ostectomy accuracy of clFHO and vFHO in dogs and found no significant difference, indicating that acceptable osteotomy can be achieved with either approach. ⁴ Although the clFHO approach limits caudal and medial visualization, the vFHO approach restricts lateral visualization, specifically of the greater trochanter and associated gluteal and rotator tendons. Damage to these structures may cause gait abnormality or in extreme cases limit weightbearing ability. It remains imperative that surgeons assess the ostectomy intraoperatively to mitigate inadvertent bone removal, even when using a surgical guide. In addition, major vessels and nerves traverse the ventral approach and must be identified and protected. Although postulated by some surgeons, ¹⁸ whether there is any clinical advantage (initial stability, pseudarthrosis, comfort, range of motion, etc) to disrupting the joint capsule and soft tissues ventrally vs craniolaterally has not been determined.

The limitations of this study include the small sample size, which may not fully represent the diversity of feline femoral morphology. We used only femurs with normal morphology and did not assess the guide’s efficacy on pathologic femurs, although many pathologies, such as slipped capital physes, should not affect guide use. We also did not compare ostectomies made via the guide with ostectomies performed freehand by a novice or experienced surgeon, or ostectomies created via a craniolateral approach. As this was an ex vivo study, we could not assess clinical outcome after vFHO using the guide. Although cadaveric evaluation suggested adequate guide positioning for ostectomy via the described approach without muscle transection, post-ostectomy soft tissue damage was not assessed; thus, no conclusions can be drawn regarding potential soft tissue damage from saw excursion.

Conclusions

This study aimed to develop a standardized ostectomy guide for vFHO in cats. The guide was practical to apply, and the actual ostectomies were statistically equivalent to the ideal ostectomies according to predefined equivalence criteria. However, we also recruited a blinded surgeon to evaluate the ostectomies in a more clinical manner, and they concluded that 5/20 specimens demonstrated either excessive (n = 3/20) or insufficient (n = 2/20) bone removal. Because of the risk of transecting the greater trochanter and compromising soft tissue attachments, the guide is not currently recommended for clinical use. Future optimization of flange size, shape and angulation may improve safety and consistency in standardized cutting guides for vFHO in cats. Additional research should be performed using more cadaveric specimens so that accurate bone removal and local soft tissue trauma associated with guided vFHO can be fully evaluated and compared with clFHO. Ultimately, a standardized guide may not prove safe for use, and customized osteotomy guides may be necessary if guided FHO is desired.