Study of digital palmar and plantar arteries of horses and mules by B-mode and Doppler ultrasonography

ABSTRACT The present study aimed to evaluate and compare the palmar and plantar digital arteries of 10 horses and 10 mules through B-mode and spectral Doppler ultrasound. Likewise, compare the plantar arteries of the pelvic limbs with the palmar arteries of the thoracic limbs, in addition to verifying the differences of the digital arteries in the contralateral limbs (right and left) in horses and mules using B-mode and spectral Doppler ultrasonography. The evaluated parameters were the diameter and the intima-media thickness (IMT), resistivity index (RI), pulsatility index (PI), peak of systolic velocity (pSV), final diastolic velocity (fDV) and mean velocity (MV) of the lateral and medial palmar and plantar arteries digital in healthy horses and mules. The horses presented higher diameters values, IMT, pSV, fDV and MV in comparison to the mules. The variables RI and PI also presented differences between horses and mules. Also, both species showed higher values in the variables of B-mode and Doppler ultrasonography in the thoracic limbs. Differences were also detected in the comparison between the limb sides (right and left) in the two modalities. The B-Mode and spectral Doppler ultrasound techniques are viable tools to identify differences in the palmar and plantar digital arteries in healthy horses and mules, highlight the differences between the thoracic and pelvic limbs and their respective sides (right and left) as well.


INTRODUCTION
The lateral and medial digital palmar arteries originate from the bifurcation of the medial artery and cross abaxially to the proximal sesamoid bones (fetlock region).The lateral and medial digital plantar arteries originate from the bifurcation of the metatarsal artery III and cross abaxially to the proximal sesamoid bones (fetlock region) (Burg et al., 2007;Parks, 2017;Fails, 2020).
Many disorders in horses result in changes in peripheral blood flow (Cochard et al., 2000), and the Doppler ultrasonography is a tool coupled with conventional ultrasonography (B-Mode) that allows vascular and hemodynamic evaluation of the vessels to be studied, enabling the diagnosis of diseases (Yanik, 2002;Souza et al., 2022).In the case of vascular alterations of digital arteries in healthy horses or with inflammatory diseases, such as laminitis and septic pododermatitis, may be detected with the aid of Mode-B and Doppler ultrasonography (Wongaumnuaykul et al., 2006;Viera et al., 2016).B-Mode ultrasonography allows the assessment of the vascular architecture (intima-media thickness (IMT)) while Doppler ultrasonography provides real-time quantitative and qualitative information on the vascular patterns, allowing an assessment of the presence or absence of flow, direction, velocity, and type of blood flow, and hemodynamic patterns (Douthit, et al, 2014;Viera et al., 2016;Miele, 2017).
Mules are hybrid animals resulting from the cross between horses and donkeys.The mules have aroused economic interest and need subsidies for scientific work to contribute to the basic knowledge to elucidate diseases, especially those of the locomotor system.Those animals are used essentially for work on farms or in gait competition, and they may not be treated as equines, since they result from the cross between equines and donkeys (Anderson, 1939;Camac, 1997;Araújo, 2010;Salles et al., 2013).
Consequently, the welfare of these animals has increased in recent years and numerous differences have been reported.Such studies were aimed at exposing possible differences and avoiding clinical, diagnostic and treatment errors (Salles et al., 2013;Miranda and Palhares, 2017;Mendoza et al., 2018).
The horses' thoracic limbs receive 60% of the animal's body weight resulting in a higher rate of illnesses when compared to the pelvic limbs (Getty, 1986;Latorre et al., 2001).For this reason, studies comparing the pelvic and thoracic limbs are required, also the improvement of the knowledge related to the hemodynamics of the blood flow of the and plantar and palmar digital arteries.
In humans, there may be differences in vessels when compared between the right and left sides (Sayer and Fatherrree, 1945), and in horses this variation may also exist Cochard et al., 2000).Studies comparing the hemodynamics of blood flow in the palmar and plantar digital arteries of equines and mules through B-mode ultrasonography and spectral Doppler were not found in the literature consulted.
In the face of the lack of information about this subject and considering the economic and social importance of these animals and the growing demand for scientific data, the present study aimed to evaluate and compare the palmar and plantar digital arteries of 10 horses and 10 mules through B-mode and spectral Doppler ultrasound.Likewise, compare the plantar arteries of the pelvic limbs with the palmar arteries of the thoracic limbs, in addition to verifying the differences of the digital arteries in the contralateral limbs (right and left) in horses and mules using B-mode and spectral Doppler ultrasonography.The criteria for selecting the animals were based on normal laboratory (blood cell count, yglutamyl transferase, aspartate aminotransferase, alkaline phosphatase, urea, and creatinine) and physical examinations (heart rate, respiratory rate, gastrointestinal motility, and rectal temperature).The animals healthy were submitted to trichotomy in the fetlock regions.

This research was approved by the Ethics
The animals were not subjected to water or solid fasting and did not receive sedatives to avoid hemodynamic changes.For B-mode and spectral Doppler ultrasonographic examinations, the animals were kept in an examination stall guided by halters and remained in stance limiting the movement of the locomotor limbs, with the body weight equally distributed between the limbs and later.Isopropyl alcohol in the concentration 70% was applied to the fetlock region for image acquisition.
The GE Healthcare Logiq V2 mobile ultrasound device with a high frequency linear transducer (8 to 12 MHz) was used to obtain B-mode images in the transverse and longitudinal plans (Fig. 1) to evaluate the diameters and IMT of the lateral and medial palmar and plantar digital artery walls.For the analysis of blood flow (spectral Doppler), insonation angles inferior to 60 degrees (60º) were used to acquire the velocity as reliably as possible.The cursor for obtaining spectral waves was positioned in the center of the lateral and medial palmar and plantar digital arteries to minimize possible changes in blood flow velocity, as recommended by Menzies-Gow and Marr (2007).
Through the spectral waves (spectral Doppler) the following variables were calculated: resistivity index (RI), pulsatility index (PI), peak systolic velocity (pSV), final diastolic velocity (fDV) and mean velocity (MV) of the lateral and medial palmar and plantar digital arteries (Fig. 2).Descriptive statistics were obtained for each group and all data were submitted to the normality test.For the statistical analysis, the Mann-Whitney test was applied to compare the two samples.Also, the Mann-Whitney test was applied to compare the measurements obtained in the pelvic and thoracic limbs, and between the contralateral limbs (right and left) in horses and mules using the SPSS V20 software.
All hypothesis tests developed in this work considered a 5% significance, with the null hypothesis being rejected when the p-value was inferior or equal to 0.05.Therefore, H0: the distributions of the two samples are equal or rejected, and H1: concludes that the two samples have significantly different distributions.

RESULTS
The lateral palmar digital artery of the right thoracic limb (RTL) showed no significant difference between horses and mules.However, the medial palmar digital artery of the RTL limb presented a difference in the longitudinal wall (p = 0.045), with horses presenting the highest mean (Table 1).
The lateral palmar digital artery of the left thoracic limb (LTL) presented a significant difference in the transverse diameter (p = 0.046), with horses presenting higher values than mules.There was a significant difference in the medial palmar digital artery of the LTL in the longitudinal diameter (p = 0.045), with the horses presenting higher values in comparison to mules (Tab.1).Table 3 shows a significant difference between horses and mules in the lateral palmar digital artery of the right thoracic limb (RTL) in the variables pSV (p <0.001), fDV (p <0.001) and MV (p <0.001), with horses showing the highest values.The medial palmar digital artery of the RTL limb presented a significant difference in the variables RI (p <0.001), PI (p <0.001), pSV (p <0.001), fDV (p <0.001) and MV (p <0.001), with higher values observed in horses in comparison with mules.
The lateral palmar digital artery of the left thoracic limb (LTL) presented a significant difference in the variables RI (p = 0.003) and PI (p = 0.029), mules with the highest means.However, for the variables pSV (p <0.001), fDV (p <0.001) and MV (p <0.001), horses presented higher values than mules.In the medial palmar digital artery of the LTL, there was a significant difference in the variables pSV (p = 0.005) and MV (p=0.038), with the highest values observed in horses.
Table 4 shows a significant difference in the lateral digital plantar artery of the right pelvic limb (RPL) in the variables pSV (p <0.001), fDV (p = 0.005) and MV (p <0.001), with horses showing the highest values.In the RPL medial plantar digital artery there was a difference in the variables PI (p = 0.004), pSV (p <0.001), fDV (p = 0.006) and MV (p = 0.002), with higher values observed in horses than mules.
The lateral and medial digital plantar artery of the left pelvic limb (LPL) presented a significant difference in the variables pSV (p <0.001), fDV (p <0.001) and MV (p <0.001), with horses showing the highest mean.
Performing the B-mode ultrasonography in horses (Table 5) there were significant differences between the pelvic and thoracic limbs in the lateral digital artery in the longitudinal (p<0.001) and transverse (p<0.001)diameters, longitudinal (p<0.001) and transversal (p<0.001)walls, whereby the thoracic limbs showed the highest values in the right and left limbs.
In the medial digital artery, there was a significant difference in the variables longitudinal (p<0.001) and transverse (p<0.001)diameters, longitudinal (p= 0.006; p<0.001) and transverse (p= 0.004; p<0.001) walls, with the thoracic limbs showing the highest means in both limbs.Performing B-mode ultrasonography in mules (Table 6), the lateral digital artery showed a significant difference between the pelvic and thoracic limbs by means of the variables: longitudinal (p<0.001) and transverse (p<0.001)diameters, longitudinal (p<0.001) and transversal (p<0.001)walls, with the left thoracic limb (LTL) presented higher value when compared to the left pelvic limb (LPL).
The medial digital artery presented difference between the pelvic and thoracic limbs in the longitudinal (p<0.001) and transverse (p<0.001)diameters, longitudinal (p= 0.007; p<0.001) and transverse (p= 0.042; p<0.001) walls, with the right and left thoracic members presenting the highest values.
Performing spectral Doppler ultrasonography in equines (Table 7), the lateral digital artery showed a significant difference between the pelvic and thoracic limbs, in the RI (p= 0.004; p<0.001),PI (p= 0.006; p<0.001) and VM (p= 0.039; p= 0.012) measures, with RI and PI showing the highest values for the right and left pelvic limbs, while VM for the right and left thoracic limbs.The medial digital artery showed a significant difference to RI (p<0.001) and PI (p<0.001) between the right pelvic limb (RPL) and the right thoracic limb (RTL), with RPL showing greater values.

Arq
The medial digital artery showed a significant difference to pSV (p= 0.006) and VM (p= 0.010) between the LPL and the LTL, in which the LTL presented the highest values.As for the fDV variable, there was a difference in the pelvic and thoracic limbs to medial digital artery, with the right (p= 0.015) and left (p= 0.002) thoracic limbs showing higher values than pelvic limbs.There were differences in RI (p= 0.027 right) and PI (p<0.001right and p= 0.011 left) between the pelvic and thoracic mules' limbs in the lateral digital artery assessed by spectral Doppler ultrasonography (Table 8), in which the pelvic limbs showed higher values than thoracic limbs.Regarding to pSV (p <0.001), fDV (p<0.001) and VM (p <0.001), there were differences between the RPL and RTL, whereby the thoracic limb presenting the highest mean.In the medial digital artery, there were difference in RI (p<0.001;p= 0.039) and PI (p<0.001;p= 0.029), in which right and left pelvic limbs presented the highest means.Related to fDV (p=0.003) and VM (p=0.017)there was a significant difference between the RPL and the RTL, with the thoracic limbs showing higher values in comparison with pelvic limbs (Table 8).Comparing the right and left thoracic horses' limbs by B-mode ultrasonography, the lateral palmar digital artery showed a significant difference in the transverse diameter (p<0.001) and in the transverse wall (p= 0.012) (Table 9), with high mean to the left limb, whereas in the medial palmar digital artery of the thoracic limbs, there was a difference only in the longitudinal diameter (p= 0.046), with LTL the highest mean.
There was no difference between the right and left limbs in the lateral digital plantar artery in the pelvic horses' limbs.However, the medial plantar digital artery showed a significant difference in the longitudinal diameter (p=0.010), with RPL showing higher mean.Comparing mules' contralateral limbs by Bmode ultrasonography (Table 10), the lateral palmar digital artery showed significant differences in the transverse diameter (p= 0.05), longitudinal wall (p= 0.026) and in the transverse wall (p = 0.049), in which LTL showed the highest value.Whilst the medial palmar digital artery showed a significant difference in the longitudinal diameter (p= 0.047) and in the transverse wall (p= 0.004), with LTL presenting the highest mean (Table 10).
There was a significant difference in longitudinal (p<0.001) and transverse (p <0.001) diameters, longitudinal (p<0.001) and transverse (p = 0.044) walls in the lateral digital plantar artery to the pelvic limbs, with RPL showing higher values than LTL.The medial plantar digital artery there was a significant difference in diameter longitudinal (p= 0,034), with RPL showing the highest values (Table 10).The lateral digital palmar artery showed difference between the right and left, pSV (p = < 0.002), fDV (p = <0.002)and VM (p = <0.012),with right showing higher values than left.The lateral digital plantar artery showed no difference between the right and left pelvic limbs, although the medial digital plantar artery showed a significant difference to RI (p= 0.039), PI (p= 0.011), pSV (p= 0.024), fDV (p= 0.04) and VM (p= 0.031), with RPL values were significantly higher in RI and PI, and LPL to pSV, fDV and VM (Table 11).

Arq
Comparing the mules' contralateral limbs by spectral Doppler ultrasonography (Tab.12), the lateral palmar digital artery showed significant differences in RI (p = 0.031), PI (p = 0.039), pSV (p = < 0.001), fDV (p = <0.001) and VM (p = <0.001),which the RI and PI were higher in the LTL, and the pSV, fDV and VM to the RTL.There was a significant difference to RI (p = 0.022), PI (p = <0.001),fDV (p <0.001) and VM (p = 0.002) in the medial palmar digital artery, with the LTL showing higher values to RI and PI, and the RTL to fDV and VM.The mules' pelvic limbs did not presented difference between the right and left limbs.

DISCUSSION
According to Torres and Jardim (1997), horses have larger hooves than the mules, which suggests that the digital arteries of the thoracic limbs are larger in horses due to the greater blood supply to the site, as confirmed ultrasound examinations B-mode and spectral Doppler in this study (Tab. 1 and 3).
We identified that the digital arteries in the pelvic limbs had higher values in the mules in some time-points (Table 2).According to Turner (2006), the horses' hooves of the pelvic limbs are smaller than those of the forelimbs, which may explain the higher values found in mules in the pelvic limbs in some measurements.Mules have intermediate characteristics between horses and donkeys (Miranda and Palhares, 2017), which may have influenced the variations in the measurements of horse and mule limbs.
Furthermore, blood vessels in bifurcating regions may present variations in diameter (Yanik, 2002) and, in addition, when evaluating B-Mode ultrasound, it is required to consider that measurements in the transverse plane may underestimate or overestimate the true values of the artery diameters and IMT (Liguori et al., 2002;Broom, 2011).However, obtaining information only through the transverse plane is not totally accurate.
In the Doppler ultrasonography, there were differences between horses and mules in the values of pSV, fDV and MV of the thoracic and pelvic limbs, which were higher in horses.This also explains the fact that the vessel diameters of horses presented higher values in almost all limbs, since the increase in blood flow is related to the diameter of the vessel (Wongaumnuayku et al., 2006;Gargano, 2015).In addition, mules had a higher mean body weight (378 kg) than horses (364 kg), which cannot rule out the possibility that body weight may have influenced the Doppler measurements.According to Kanegusuku et al. (2020), human patients with higher body weight tend to have lower blood flow velocity, which may explain the fact that horses had higher values in the pSV, fDV and MV variables.Furthermore, some studies have shown that, despite considerable variability in measurements, the use of Doppler ultrasonography in the equine limb is sufficiently repeatable and sensitive to detect small changes in blood flow (Raisis et al., 2000;Menzies-gow;Marr, 2007;Douthit et al., 2014).
The variables RI and PI presented variations between horses and mules in this study.According to the literature, the mules are hybrid equines (Anderson, 1939;Camac, 1997;Araújo, 2010;Salles et al., 2013) and, therefore, present differences in comparison to the anatomophysiological studies of horses (Burnhan, 2002;Alsafy et al., 2008;Smith, 2009;Miranda and Palhares, 2017) which may also explain the differences observed in this study.It is also essential to emphasize that the RI variable may be influenced by several factors, such as the compliance of the vessel wall, blood pressure and long-term effects of changing the hull geometry.
In horses and mules, the thoracic limbs are usually more affected by locomotor diseases; however, they may occur in all four limbs (Kahn, 2008).The digital palmar arteries of the thoracic limbs of horses and mules showed higher values of diameter and IMT when compared to those of the pelvic limbs.In spectral Doppler ultrasonography, horses and mules had higher values of pSV, fDV and VM in the thoracic limbs.According to Gargano (2015), blood flow is directly related to the increase in arterial diameter, which justifies the fact that horses and mules presented higher values of diameters and IMT (B-mode) for the thoracic limbs.Blood flow and vessel diameters were higher in the thoracic limbs, and naturally, the blood supply to these limbs is greater (Butler et al., 2017) which may be related to the fact that those limbs support 55% to 60% of the animal's entire body weight, thus the blood flow demand for the digit must be proportional to the weight supported (Dyce et al., 1997).Additionally, when a vessel undergoes increased resistance (RI), blood flow undergoes a smaller area of constant flow (Cipone et al., 1997;Yanik, 2002;Varughese et al., 2013), observed in the present study in horses and mules pelvic limbs.
In humans, blood vessels may present differences between the right and left limbs (Sayer and Fatherrree, 1945), also observed in the present study in both horses and mules.The arteries of the upper humans' limbs can vary from 9 to 24% (Celik et al,. 2001).In animals, this difference can also be observed, and this greater asymmetry is observed in the locomotor members (Wood, 1997).
horses presented no difference in the lateral palmar digital artery and lateral plantar digital artery between the contralateral limbs by Doppler ultrasonography, similar to the observed by Menzies-Gow and Marr (2007) and Aguirre et al. (2013), who evaluated the lateral palmar Arq.Bras.Med.Vet.Zootec., v.75, n.4, p.571-590, 2023 digital artery of both thoracic limbs of horses.However, the horses showed differences between the contralateral limbs in the medial palmar digital artery of the thoracic limbs and medial digital plantar artery of the pelvic members.Whilst in the mules, there were differences only between the thoracic limbs in the lateral and medial palmar digital arteries, in the pelvic limbs no difference was observed.According to Cochard et al. (2000), the comparison of blood flow in the digital palmar arteries of the horses' thoracic limbs using Doppler ultrasonography should not present significant differences between the contralateral limbs, although there may be some variability, different what was observed in the present study.
Only horses showed differences in the medial digital plantar arteries between the pelvic limbs.However, in healthy cattle Müller et al. (2017), observed difference between the right and left pelvic limbs in the lateral and medial digital plantar arteries by spectral Doppler ultrasonography.According to Stashak and Hill (2006), the animal's weight is not evenly distributed when resting its limb on the ground, supporting more medial lower of the hoof.The differences in changes in hoof pressure, induced by weight changes, may be responsible for the difference in blood flow between the right and left sides of the horses' locomotor limbs (Raisis et al., 2000;Stashak andHill, 2006, Müller et al., 2017).
Depending on the anatomy, the lateral and medial palmar digital arteries are responsible for the digit's arterial supply, which originate from the bifurcation of the medial palmar artery and the lateral and medial digital plantar arteries originate from the bifurcation of the dorsal metatarsal artery III (Burg et al., 2007;Parks, 2017;Fails, 2020).
According to Yanik (2002) changes in blood flow, vessel sizes and viscosity in regions of vessel bifurcations can occur, in addition to biological factors may change the blood flow (Cochard et al., 2000;Menzies-Gow and Marr, 2007) which can justify the differences found between the contralateral in horses and mules' limbs.
The limb movement during B-mode and spectral Doppler ultrasonography exams is another factor that can influence the digital arteries assessment, which can result in variations in blood flow velocity and the resistivity index (RI) (Menzies-Gow and Marr, 2007).Hoffmann et al. (2001), using spectral Doppler to analyze the animals' lateral palmar digital arteries with weight support (load) and weightlessness (no load), observed variations between 23.6 to 37.8% of the blood flow velocity in the limbs.Wongaumnuaykul et al. (2006), evaluated medial palmar digital arteries by spectral Doppler ultrasonography and observed variations in blood flow velocity (8.1 to 19.8%) between the control groups, the group with septic pododermatitis and laminitis.In the present study, ultrasonography examinations (Bmode and Doppler) were performed with body weight equally distributed over the four limbs to avoid hemodynamic changes and the RI, trying to maintain a pattern and minimum animals' mobility.
Despite B-mode and Doppler ultrasonography are influenced by technical and biological factors, there are studies that describe the use of this tool in diagnosis (Hunt et al., 1994;Ingle-Fehr and Baxter, 1998;Cochard et al., 2000;Menzies-Gow and Marr, 2007) and in the treatments monitoring, such as septic pododermatitis and laminitis in horses (Wongaumnuaykul et al., 2006).

CONCLUSION
Digital arteries had higher diameter and IMT values in horses in almost all limbs (B-Mode ultrasound).As for spectral Doppler, horses presented higher values in the variables pSV, fDV and MV, while RI and PI presented differences between horses and mules.
Horses and mules had the values of diameter, IMT, and blood flow superior to the thoracic limbs and differences between the contralateral limbs.The modalities of B-mode and spectral Doppler ultrasonography demonstrated efficacy in the evaluation of the palmar and plantar digital arteries in horses and mules and may show differences between the thoracic and pelvic limbs and their respective sides, and between animals, emphasizing the importance of performing ultrasonography in healthy animals.
ultrasonography and spectral Doppler were performed on the lateral and medial digital palmar and plantar arteries of 10 healthy horses (5 males and 5 females), with body weight between 336 and 390 kg and ages between 2 and 18 years.The examinations were also performed in the same anatomical regions in 10 mules (5 males and 5 females), with body weight between Study of digital… Arq.Bras.Med.Vet.Zootec., v.75, n.4, p.571-590, 2023 573 347 and 439 kilograms (kg) and age between 4 and 15 years.

Figure 1 .
Figure 1.Position of the transducer to obtain images in the planes: transverse (A -horse) and longitudinal (C -mule).B-mode ultrasound image of the digital artery in the planes: transverse (B) and longitudinal (D).

Figure 2 .
Figure 2. Position of the transducer in the longitudinal plane to perform spectral Doppler ultrasonography (A -horse).Post-processing of images with the contour of the waves to obtain the values of RI and PI (B), and pSV, fDV and MV (C).

Table 1 .
Mean, median, standard deviation, Q1, Q3 and confidence interval of the variables in B-mode ultrasonography of the lateral and medial digital arteries of the thoracic limbs, for comparison between horses and mules

Arq. Bras. Med. Vet. Zootec., v.75, n.4, p.571-590, 2023Table 2 .
Mean, median, standard deviation, Q1 and Q3 of the variables in B-Mode ultrasonography of the lateral and medial digital arteries of the pelvic limbs, for comparison between horses and mules

Table 3 .
Mean, median, standard deviation, Q1 and Q3 of the variables of spectral Doppler ultrasonography of the lateral and medial digital arteries of the thoracic limbs, for comparison between horses and mules

Table 4 .
Mean, median, standard deviation, Q1 and Q3 of the variables of the spectral Doppler ultrasonography of the lateral and medial digital arteries of the pelvic limbs, for comparison between

Table 5 .
Mean, median, standard deviation, Q1 and Q3 of the horses' limbs variables by B-mode ultrasonography, followed by the p-value to compare the digital arteries between the pelvic and thoracic limbs

Table 6 .
Mean, median, standard deviation, Q1 and Q3 of the mules' limbs variables by B-mode ultrasonography, followed by the p-value to compare the digital arteries between the pelvic and thoracic limbs

Table 8 .
Mean, median, standard deviation, Q1 and Q3 of the mules' limbs variables by spectral Doppler ultrasonography, followed by the p-value to compare the digital arteries between the pelvic and thoracic

Table 9 .
Mean, median, standard deviation, Q1 and Q3 of the horses' limbs variables by B-mode ultrasonography, followed by the p-value to compare the digital arteries between the pelvic and thoracic limbs

Table 10 .
Mean, median, standard deviation, Q1 and Q3 of the mules' limbs variables by B-mode ultrasonography, followed by the p-value to compare the digital arteries between right and left side of pelvic and thoracic limbs

Table 11 .
Mean, median, standard deviation, Q1 and Q3 of the horses' limbs variables by spectral Doppler ultrasonography, followed by the p-value to compare the digital arteries between right and left side of pelvic and thoracic limbs

Table 12 .
Mean, median, standard deviation, Q1 and Q3 of the mules' limbs variables by spectral Doppler ultrasonography, followed by the p-value to compare the digital arteries between right and left side of pelvic and thoracic limbs