Radiographic patterns and lesion associations in feline pelvic trauma

Abstract

Objectives

Pelvic fractures are common in feline trauma patients and often involve multiple structures due to the rigid anatomy of the pelvis. This study aimed to investigate the radiographic distribution of pelvic lesions in cats, to identify potential correlations among specific pelvic lesions and to evaluate the relationship between lesion type and total number of lesions.

Methods

A retrospective bicentric study was conducted on cats presenting with pelvic trauma, based on orthogonal pelvic radiographs. Pelvic lesions were classified anatomically, and statistical analyses were performed to evaluate associations among different lesion types and between lesion type and total number of lesions.

Results

A total of 80 cats with pelvic trauma were included. A total of 267 pelvic lesions were identified, with a mean of three lesions per patient, and most cats (91.2%) had multiple pelvic lesions. Sacroiliac luxation was the most common lesion. Positive statistically significant associations were identified between (1) left and right sacroiliac luxation (ρ = 0.20, P = 0.036), (2) sacroiliac luxation with contralateral iliac fracture (ρ = 0.21, P = 0.030; ρ = 0.24, P = 0.017) and (3) pubic fractures with ipsilateral ischial fractures (ρ = 0.53, P <0.001; ρ = 0.37, P <0.001). Unilateral sacroiliac luxation, pubic, ischial and pelvic symphyseal fractures were never observed in isolation, whereas coxofemoral luxation often occurred as a solitary lesion.

Conclusions and relevance

Feline pelvic trauma frequently results in complex, multi-site lesion patterns, and consistent associations among specific lesions suggest predictable biomechanical behaviour within the pelvic ring. Recognition of these patterns may assist clinicians in the systematic radiographic evaluation of feline pelvic trauma and facilitate detection of additional, potentially occult lesions.

Plain language summary

Pelvic injuries in cats: how common they are and which lesion combinations veterinarians should expect on radiographs

Pelvic fractures are a common consequence of accidents in cats, especially when they experience high-energy trauma. The pelvis is a rigid structure, so when a cat experiences a significant impact, the force usually spreads through the whole pelvic ring, creating multiple lesions rather than a single injury. This study examined radiographs from 80 cats with pelvic trauma to understand how pelvic lesions are distributed and whether certain injuries tend to occur together. Most cats (over 90%) had lesions affecting more than one part of the pelvis. On average, each cat had three separate pelvic lesions. The most common lesion was sacroiliac luxation, a condition in which the pelvis becomes separated from the spine on one or both sides. Several injuries were found to occur together, suggesting predictable trauma patterns. For example, sacroiliac luxation was often associated with a fracture of the ilium on the opposite side, whereas pubic fractures commonly appeared together with ischial fractures on the same side. These associations likely reflect how forces travel through the pelvis during trauma. Some lesions, such as fractures of the pubis, ischium, pelvis midline (symphysis) or sacroiliac luxation, rarely occurred on their own and were usually part of more complex injury patterns. Conversely, hip dislocation (coxofemoral luxation) was more likely to appear as an isolated lesion. By highlighting these patterns, this study provides practical guidance to veterinarians. Recognising which lesions commonly occur together may help clinicians to interpret radiographs more accurately and detect additional lesions that might otherwise be missed.

Introduction

Pelvic fractures are a frequent outcome of high-energy trauma in cats, accounting for approximately 20–32% of all feline fractures documented in the veterinary literature.^{1 –8}

These lesions are most commonly reported in young, male cats with outdoor access, likely due to their increased exposure to environmental risk and their limited experience in responding to unfamiliar threats.^3,4,9 Among the causes of pelvic trauma, road traffic accidents are consistently reported as the leading cause, followed by falls from heights (high rise syndrome), crush injuries and, less frequently, bite wounds or human-inflicted trauma.^{1,3 –6,8 –11}

The rigid, box-like structure of the feline pelvis makes it highly resistant to damage, typically requiring substantial traumatic force to fracture.^{3,12 –15} Consequently, these lesions are rarely isolated and often involve multiple pelvic bones, such as the ilium, pubis, ischium and acetabulum, in various combinations.^1,5,7,13,16 In addition, up to 94% of affected cats can present with concurrent orthopaedic or soft tissue injuries, including neurological damage or urinary tract disruption.^{1,5,7,12 –14,17}

Diagnostic imaging plays a crucial role in evaluating pelvic injuries. ¹⁸ Radiography remains the first-line imaging modality and is widely available, although it can miss subtle or complex fracture patterns due to bone and soft tissue overlap.^{2,8,12,14,17,19} In comparison, CT provides superior anatomical detail but is more expensive and less readily available in veterinary medicine, with no clear evidence of a substantial impact on treatment decisions.^{19 –22}

This study aimed to investigate the radiographic distribution of pelvic lesions in cats and to identify potential associations among different pelvic lesions. We hypothesised that specific pelvic lesions may occur in consistent association patterns as a result of force transmission within the pelvic ring.

Materials and methods

This retrospective bicentric study included feline patients referred to a university veterinary teaching hospital and a private veterinary trauma referral clinic. Both client-owned and stray cats were eligible for inclusion.

Inclusion criteria were a documented history of trauma or, in the absence of such history, clinical suspicion of traumatic injury. Eligible patients were also required to have had two orthogonal pelvic radiographs where there was evidence of at least one pelvic lesion.

Pelvic lesions were classified as sacroiliac luxation, coxofemoral luxation or fractures involving the ilium, ischium, pubis, acetabulum, ischiatic tuberosity and pelvic symphysis. Sacroiliac joint injuries were recorded under the single category of ‘sacroiliac luxation’. No distinction was made between complete and partial displacement, as standardised radiographic criteria differentiating luxation from subluxation in cats are not currently established. For this reason, any radiographic evidence of sacroiliac joint malalignment or displacement was classified as luxation, regardless of the degree of displacement. All lesions, except for those involving the pelvic symphysis, were recorded as right- or left-sided. The anatomical distribution and the total number of pelvic lesions were recorded.

Signalment data, including age, sex and breed, were recorded when available. Because microchipping is legally mandatory in the region of the participating institutions, cats without a microchip or with a microchip registered to the regional health authority (eg, cats belonging to registered feline colonies) were considered stray.

Because stray cats were often presented without documented history, complete signalment information was not always available. In such cases, age was estimated based on clinical examination and radiographic indicators of skeletal maturity. Male cats were classified as intact or castrated at presentation, whereas neuter status in female patients could not always be definitively determined without additional diagnostic procedures; when this information was not explicitly documented in the medical record, female cats were classified as intact.

Exclusion criteria included the presence of pathological fractures and radiographs of inadequate diagnostic quality.

Radiographs were evaluated by consensus between a veterinary graduate PhD student and a board-certified specialist in veterinary diagnostic imaging.

Statistical analysis was performed using SPSS version 28 (IBM). Age distribution was assessed for normality and found to be non-normally distributed; therefore, age is presented as median 25th and 75th percentiles (25th–75th). Categorical variables, including sex and breed, were expressed as frequencies and percentages.

Associations between lesion types and total number of lesions were evaluated using the χ² test. Statistical significance was set at P <0.05. Correlations between different lesion types were assessed using Spearman’s correlation coefficients to identify possible patterns among specific lesion types. The Spearman’s rho (ρ) coefficient was interpreted as follows: less than 0.10 = negligible correlation; 0.10–0.39 = weak correlation; 0.40–0.69 = moderate correlation; 0.70–0.89 = strong correlation; and 0.90–1.00 = very strong correlation. ²³ A correlation was considered statistically significant when P <0.05.

Results

A retrospective analysis of the databases from the two selected institutions initially identified 84 feline patients. Of these, four were excluded because of either the presence of pathological fractures (n = 1/4) or inadequate radiographic quality (n = 3/4).

The final study population included 80 cats that had experienced trauma involving the pelvic region. Among these, 50 were stray cats presented to a university veterinary teaching hospital intensive care unit, while the remaining 30 were owned cats treated at a private veterinary referral clinic.

The study population had a median age of 29 months (interquartile range 12–60) and included 30% intact males, 20% castrated males, 32.5% intact females and 17.5% spayed females. The majority of cats were domestic shorthairs (95%), with only a small representation of Bengal (1.3%), Norwegian Forest (2.5%) and Persian cats (1.3%).

A total of 267 pelvic lesions were identified, with a mean of 3.33 lesions per patient (range 1–6).

Three lesions per patient were the most frequent finding, occurring in 21/80 (26.2%) cases, followed by two (22.5%), four (21.3%), five (11.2%) and six lesions (10.0%). Only seven (8.8%) cats had a single pelvic lesion.

The most commonly observed lesions were right sacroiliac luxation (n = 41/80, 51.3%) and left sacroiliac luxation (n = 37/80, 46.3%). Of the 80 cats included in the study, 23 (28.8%) were diagnosed with bilateral sacroiliac luxation. Only one (4.3%) of these cases had bilateral sacroiliac luxation as an isolated finding. The remaining 22/23 (95.7%) cases had additional pelvic lesions. When associated lesions were considered individually, pelvic symphyseal fracture was the most frequently observed concurrent finding (n = 9/23, 39.1%), followed by left pubic fractures (n = 7/23, 30.4%). Left ischial fractures and right pubic fractures were each identified in 6/23 (26.1%) cases, while iliac fractures (either left or right) were present in 5/23 (21.7%) cases. Right ischial fractures were observed in 4/23 (17.4%) cases and left acetabular fractures in 2/23 (8.7%) cases. Isolated cases (n = 1/23, 4.3% each) of ischial tuberosity fracture and coxofemoral luxation (right or left) were also recorded. No cases of bilateral sacroiliac luxation were associated with right acetabular fracture.

Other frequent lesions included fractures of the pubis (right and left) and pelvic symphyseal fractures (n = 27/80, 33.8%), followed by fractures of the ilium (right: n = 19/80, 23.8%; left: n = 16/80, 20%) and ischium (left: n = 21/80, 26.3%; right: n = 14/80, 17.5%). Less common lesions were left coxofemoral luxation (n = 10/80, 12.5%) and left acetabular fracture (n = 9/80, 11.3%). The least frequent lesions were fractures of the ischiatic tuberosity (right and left) and right acetabular fracture (n = 5/80 each, 6.3%) and right coxofemoral luxation (n = 4/80, 5.0%).

Spearman’s correlation analysis revealed several statistically significant associations among specific pelvic lesions (Table 1). Positive correlations were observed between right and left sacroiliac luxation (ρ = 0.20, P = 0.036), sacroiliac luxation and contralateral iliac fracture (both right and left, ρ = 0.21, P = 0.030; ρ = 0.24, P = 0.017) and left sacroiliac luxation with pelvic symphyseal fracture (ρ = 0.19, P = 0.049).

Table 1

Spearman’s correlation analysis of positive and negative associations among specific pelvic lesions, with the number of cats presenting each evaluated association

Association	Spearman’s rho	P value	Cats (n = 80)
Left sacroiliac luxation – right sacroiliac luxation	0.203	0.036	23
Left sacroiliac luxation – right ilium	0.211	0.030	12
Left sacroiliac luxation – pelvic symphysis	0.186	0.049	16
Left sacroiliac luxation – left coxofemoral luxation	–0.199	0.038	2
Right sacroiliac luxation – left ilium	0.238	0.017	12
Right sacroiliac luxation – left coxofemoral luxation	–0.236	0.017	2
Left ilium – right ilium	–0.269	0.008	0
Left ilium – right pubis	0.304	0.003	10
Right ilium – right pubis	–0.259	0.010	3
Right ilium – left coxofemoral luxation	–0.203	0.035	0
Left pubis – left ischium	0.535	<0.001	16
Left pubis – left coxofemoral luxation	–0.190	0.046	1
Right pubis – right ischium	0.367	<0.001	11
Left ischium – left coxofemoral luxation	–0.225	0.022	0
Right ischium – pelvic symphysis	–0.190	0.046	2
Left ischiatic tuberosity – left coxofemoral luxation	0.260	0.010	2

Several ipsilateral associations were also noted, including pubic and ischial fractures (both left and right, ρ = 0.53, P <0.001; ρ = 0.37, P <0.001), and left ischiatic tuberosity fracture with left coxofemoral luxation (ρ = 0.26, P = 0.010) (Figure 1).

Figure 1

Examples of statistically significant positive associations between specific pelvic lesions. (a) Ventrodorsal radiograph of a young adult female domestic shorthair cat (approximately 1 year old). Bilateral sacroiliac luxations are present (solid arrows). A left ischiatic tuberosity fracture is also visible (arrowhead). (b) Ventrodorsal radiograph of an adult spayed female domestic shorthair cat. Left pubic and left ischial fractures (arrowheads) are present, with a concurrent right sacroiliac luxation (solid arrow). (c) Ventrodorsal radiograph of a young female domestic shorthair cat (approximately 6 months old). A right iliac fracture (solid arrow) is present with contralateral left sacroiliac luxation (dashed arrow). A fracture of the pelvic symphysis is present but not clearly identifiable in this projection because of superimposition; the lesion was confirmed on the lateral view. A left ischiatic tuberosity fracture through the physis is also present (arrowhead). The contralateral ischiatic tuberosity physeal line is visible, consistent with skeletal immaturity, but without displacement

Negative correlations included left and right sacroiliac luxation with left coxofemoral luxation (ρ = –0.20, P = 0.038; ρ = –0.24, P = 0.017), right iliac fracture with left iliac or right pubic fracture (ρ = –0.27, P = 0.008; ρ = –0.26, P = 0.010), left coxofemoral luxation with left pubic and left ischial fractures (ρ = –0.19, P = 0.046; ρ = –0.22, P = 0.022), and right ischial fracture with pelvic symphyseal fracture (ρ = –0.19, P = 0.046).

Finally, the relationship between the total number of lesions per patient and the presence of specific pelvic lesions was assessed. Sacroiliac luxation, pubic fractures, left ischial fractures and pelvic symphyseal fractures were never found as isolated lesions and showed increased prevalence with higher number of total lesions, although the trend varied slightly across lesion types.

Conversely, left coxofemoral luxation exhibited the opposite pattern: it was most frequently observed as a single lesion (57.1%) and decreased in frequency with increasing number of total lesions (Figure 2).

Figure 2

Decision-support algorithm for the radiographic evaluation of feline pelvic trauma. The flowchart summarises the prevalence of the pelvic lesions identified in this study and the significant positive and negative correlations among lesion types

Discussion

This study investigated patterns of pelvic lesions in a population of cats with pelvic trauma, aiming to identify potential associations among different pelvic lesions.

The study population was predominantly composed of domestic shorthair cats with a balanced sex distribution and a median age of 29 months, which is consistent with previous reports describing pelvic trauma as more frequently affecting young adult cats.^1,5,9,24,25 This distribution may reflect behavioural characteristics typical of younger animals, such as increased activity and exploratory behaviour, which may predispose them to traumatic events. The predominance of domestic shorthair cats likely reflects the general composition of the feline population, particularly among stray cats, rather than indicating a breed predisposition.

Our results revealed a high prevalence of multi-site pelvic lesions, with a mean of more than three lesions per patient and only 8.8% of cats presenting with a single lesion. This finding is consistent with previous reports, which describe the feline pelvis as a rigid, box-like structure predisposed to widespread injury due to energy redistribution during high-impact trauma.^1,7,13,14,16

Sacroiliac luxation, both left and right, was the most frequently observed lesion, followed by fractures of the pubis and pelvic symphysis. This is partly consistent with previous studies, where sacroiliac luxation has been identified as the most common pelvic lesion, with reported prevalence as high as 90%.^5,16,24,26 However, some studies have found pelvic floor injuries to be more prevalent. ¹

In addition to differences in overall prevalence, previous studies have also reported variability in the proportion of unilateral vs bilateral sacroiliac luxations. In the present study, bilateral sacroiliac luxation was identified in 28.8% of cats, while earlier reports described sacroiliac luxation as predominantly unilateral, with bilateral involvement reported less frequently. ²⁴

These discrepancies in reported prevalence may partly reflect differences in the classification and detection of sacroiliac joint lesions. In particular, subtle or partial displacement of the sacroiliac joint may be difficult to recognise on standard radiographs and could therefore be underdiagnosed, potentially influencing the reported prevalence of pelvic lesions. In the present study, sacroiliac joint lesions were analysed as a single entity, consistent with the existing veterinary literature. Indeed, most published reports describe sacroiliac ‘luxation’ as a single diagnostic category, without standardised radiographic criteria distinguishing complete from partial joint displacement. Even recent studies proposing objective radiographic measurements have focused primarily on differentiating normal from abnormal sacroiliac joints rather than grading the severity of joint malalignment.^11,27

The incidence of pelvic symphyseal fractures in our study was comparable to that reported by Yurtal et al, ¹⁵ suggesting this may represent a relatively frequent injury pattern, although other colleagues characterise it as rare. ²⁸

The frequency of iliac fractures in our population was also similar to that reported by Yurtal et al, ¹⁵ with iliac involvement noted in approximately 20% of cases. Conversely, acetabular fractures were among the least frequent lesions in this cohort, consistent with other reports. ²⁰ This type of fracture may be underdiagnosed on radiographs due to superimposition and complex anatomy; indeed, CT has been shown to be more sensitive for detecting acetabular lesions.^{19 –21}

Several positive associations were identified between specific pelvic lesions, including simultaneous left and right sacroiliac luxation, sacroiliac luxation with contralateral iliac fracture and pubic fractures with ipsilateral ischial fractures. These consistent associations suggest the presence of recognisable injury patterns and emphasise the importance of thorough radiological evaluation when certain lesions are identified. Specifically, the associations between sacroiliac luxation and contralateral iliac fracture, as well as bilateral sacroiliac luxation, have been previously reported and are thought to reflect force transmission through the pelvic ring.^1,26

In contrast, negative correlations, such as between sacroiliac and ipsilateral coxofemoral luxation or between right and left iliac fractures, may be the result of biomechanical ‘protection’ through force dissipation. Similar protective patterns have been noted in earlier studies, including the inverse relationship between sacroiliac and ipsilateral coxofemoral luxation. ¹

Another relevant finding concerns the tendency for certain lesions, such as sacroiliac luxation, pubic, ischial and pelvic symphyseal fractures, to occur in combination rather than in isolation. This may reflect the high-energy nature of pelvic trauma, in which forces propagate through the pelvic ring, resulting in involvement of multiple anatomical regions and complex, multi-site injury patterns.^1,7,13,14,16

Clinically, recognition of these patterns is crucial and should prompt comprehensive evaluation for additional, potentially occult, fractures.

In contrast, coxofemoral luxation frequently occurred as a solitary lesion. This may reflect a different injury mechanism, possibly involving lower-energy trauma or a localised force centred at the hip joint or pelvic limb rather than the entire pelvis.

This study has several limitations. First, although age was reported descriptively, skeletal maturity was not systematically assessed and no separate analysis was performed according to age group; therefore, potential age-related differences in pelvic lesion patterns were not specifically evaluated. Second, the presence of concomitant soft tissue or extra-pelvic skeletal injuries was not considered in the present study. This is particularly relevant in polytraumatised patients, in which comorbidities involving the spine or urinary tract are common and clinically significant.^1,5,7,12,13 Third, because of the retrospective nature of the study, diagnostic accuracy depended on the quality of the available radiographs. Although radiography is generally considered sensitive for detecting pelvic lesions,^2,14,18,19 superimposition of anatomical structures may have led to underdiagnosis in some cases. In addition, because standardised radiographic criteria to differentiate sacroiliac luxation from subluxation are not currently established in cats, these lesions were not stratified according to degree of displacement, which may have limited assessment of potential correlations related to the severity of sacroiliac joint displacement.

Conclusions

This study highlights the high prevalence and complexity of pelvic lesions in traumatised cats, with most patients sustaining multiple concurrent lesions. Consistent patterns of concurrent lesions suggest a degree of biomechanical predictability in feline pelvic trauma. In particular, the identification of sacroiliac luxation, pubic fracture, left ischial fracture or pelvic symphyseal fracture should prompt thorough investigation for additional lesions, as these rarely occur in isolation. These findings may assist in the interpretation of pelvic radiographs, especially for less experienced clinicians, and underscore the importance of careful imaging assessment and lesion pattern recognition in the management of feline pelvic trauma. Further research involving larger populations and the use of advanced imaging modalities (eg, CT) is needed to further refine fracture pattern prediction and evaluate associations with additional skeletal or soft tissue injuries.