Thickening of the patellar tendon, termed patellar tendinopathy, patellar tendinosis, or patellar tendinitis, is a common postoperative complication in dogs following TPLO for treatment of cranial cruciate ligament injuries. The condition in dogs has been poorly characterized on the cellular level but is likely most similar to the condition of patellar tendinosis described in human patients.1 For the sake of clarity, the condition in dogs will be referred to as a tendinopathy, whereas references to the human condition will be termed tendinosis, which is a term commonly used in the human literature. Between 80% and 100% of dogs undergoing TPLO develop radiographically or ultrasonographically detectable abnormal thickening of the distal portion of the patellar tendon.1,2 Clinical signs of focal pain and lameness develop in up to 25% of cases with patellar tendon thickening.1 Ultrasonographically detectable changes in thickness of the distal portion of the patellar tendon are evident as early as 4 weeks after TPLO.2 The diagnosis of patellar tendinopathy in dogs is based on a thickened distal portion of the patellar tendon (palpation of which elicits signs of pain) or identification of thickening of the distal portion of the patellar tendon on routine postoperative radiographs. Additionally, the authors have observed this condition in some dogs after tibial tuberosity transposition as a treatment for grade IV medial patella luxation, although similar reports of patellar tendon swelling after tibial tuberosity transposition could not be found in the literature.
The cause of postoperative patellar tendinopathy is not well understood. Speculated causes include increased patellar tendon strain after TPLO, thermal trauma during osteotomy, trauma secondary to Kirschner wire placement through the insertion site of the distal portion of the patellar tendon, large alterations in the tibial plateau angle, large movements of the intercondylar eminences during tibial plateau rotation, and excess postoperative activity.1–3 The role of increased patellar tendon strain after TPLO is controversial. Investigators of 2 studiesa,b found differing results regarding the effect of TPLO on patellar tendon strain. In 1 study,a investigators found no difference, whereas investigators of the other studyb found a 24% increase in patellar tendon strain. Clinical investigations have failed to definitively correlate any of these factors with the development of patellar tendinopathy.1–3 It remains unclear whether postoperative patellar tendinopathy is caused by chronic changes within the tendon after surgical trauma or is secondary to adaptive structural changes resulting from surgical alteration of stifle joint conformation (or is a combination of both).
Arthrotomy is a procedure for examination and exploration of the stifle joint and is commonly used to treat a variety of injuries or conditions. Stifle joint arthrotomies are often extended distally to the level of the tibial plateau and proximally to the level of the patella and through the femoropatellar ligament to increase exposure of intra-articular structures. These surgical interventions may combine to substantially reduce the blood supply to the patellar tendon. If there is local disruption of blood delivery, effects of acute ischemia on the ability of the tendon of dogs to heal or to adapt to alterations in biomechanical strain during the postoperative period are not known.
The purpose of the study reported here was to determine whether arthrotomy alone or in combination with osteotomy of the proximal portion of the tibia would affect blood delivery to the patellar tendon. Additionally, the relative numbers of contrast-infused vessels were compared within various portions of the patellar tendon in the control and treatment groups to determine whether any areas with detected impairment of blood delivery could be correlated with the location of clinical patellar tendinopathy. The null hypothesis was that the various surgical techniques would have no effect on blood delivery to the patellar tendon.
Materials and Methods
Sample
Twenty-four cadavers of mixed-breed dogs euthanized for reasons unrelated to the present study were obtained from a local animal shelter. Cadavers of medium- to large-breed nonchondrodystrophoid dogs that weighed between 20 and 40 kg were obtained and stored at −20°C until use. Immediately before use, cadavers were thawed in a warm water bath and examined to rule out relevant orthopedic disease, especially that involving the stifle joint. For evaluations involving contrast agent, joints were radiographically examined to detect changes consistent with osteoarthritis.
Treatment groups
One stifle joint from each cadaver was assigned to 1 of 4 treatment groups: MA group, LA group, MA and LA with tibial tuberosity transposition (MALA group), and MA with TPLO (TPLO group). There were 6 limbs/treatment group. The contralateral limb served as the nonoperated control sample. Each cadaver was assigned by use of a randomization procedure that involved preassigning procedures to OR tables within a multiple-station laboratory area; cadavers were then placed on the tables as they were removed from the cadaver bags. Individual limbs of each cadaver were assigned to control or treatment groups to ensure that each group had an equal number of right and left limbs.
Surgical procedures
All surgical interventions were performed by 1 investigator (MDJ). For the MA and LA groups, a parapatellar incision was made approximately 1 cm medial or lateral to the patellar tendon beginning at the level of the proximal portion of the tibia and extending proximally to include incision through the femoropatellar ligament to the level of the proximal portion of the patella.4 For the groups undergoing osteotomy of the proximal portion of the tibia (MALA and TPLO groups), the appropriate arthrotomy skin incision was extended distally to allow exposure of the necessary region of the tibia. The infrapatellar fat pad was preserved during extension of the arthrotomy to preserve its contribution to the patellar tendon blood supply.
For the MALA group, MA and LA incisions were performed as described previously and were extended distally to expose the tibial tuberosity. Periosteum of the tibial tuberosity was elevated medially and laterally in the region of the osteotomy. The osteotomy was performed with a handheld osteotome and mallet in a proximal-to-distal direction. The osteotomy was placed immediately cranial to the muscular groove of the tibia and immediately caudal to the infrapatellar fat pad.5 Care was taken to minimize trauma to the infrapatellar fat pad and preserve its contribution to the patellar tendon blood supply. The osteotomy was continued until the tibial tuberosity could be moved laterally while leaving the distal aspect of the periosteum intact. The tibial tuberosity segment was moved laterally 2 to 4 mm and stabilized to the proximal portion of the tibia with a 1.0-mm Kirschner wire that simulated the surgical procedure used for patellar luxation.
For the TPLO group, the MA was extended distally over the proximal aspect of the tibia to include the tendinous attachment of the semitendinosus muscle. The periosteum was elevated between the medial collateral ligament and tibial tuberosity. The popliteus muscle was elevated from the caudal aspect of the proximal portion of the tibia. The lateral aspect of the tibial tuberosity was not surgically approached. Biradial osteotomy for the TPLO was placed as caudally and proximally as possible. This location preserved as much of the tibial tuberosity as possible and placed the proximal aspect of the osteotomy immediately cranial to the tibial plateau and caudal to the infrapatellar fat pad. Preoperative radiographs were not obtained, so calculations of the tibial plateau angle were not made before TPLO. After the biradial osteotomy was made, the tibial plateau segment was rotated a representative amount (approx 5 to 10 mm) and secured with a 1.0-mm Kirschner pin placed through the proximal portion of the tibial tuberosity into the rotated segment of the tibia. The Kirschner pin was placed through the patellar tendon in a similar manner to that for clinical situations to temporarily hold the rotated segment in place until the medial plate could be applied.
Evaluation involving contrast agent
A ventral midline abdominal incision was made in each cadaver. An 8F feeding tube was used as a catheter and inserted into the abdominal aorta at a location caudal to the renal arteries but cranial to the trifurcation of the iliac arteries. The aorta was double ligated with 2–0 polypropylene at a point cranial to the tube to prevent backflow of contrast agent. After the catheterization was completed but before there was any surgical intervention, a large volume of saline (0.9% NaCl) solution was infused through the catheter until clear effluent was visible in the caudal vena cava. The location for the catheter allowed simultaneous injection of contrast agent into both hind limbs (treatment and control) and enabled inclusion of all potential vascular contributors to the patellar tendon. Several studies6–10 have found good results for microangiographic analysis with barium sulfate solutions (range, 30% to 62% [wt/vol]); however, because of the large volume of tissue requiring perfusion, radiopaque contrast material (barium sulfate) was mixed with saline solution to reduce viscosity. India ink was added as a marker for histologic analysis. The final solution consisted of 33% barium, 17% India ink, and 50% saline solution.
The surgical procedure was performed on all dogs prior to injection of contrast agent. Additionally, before injection of the contrast solution, each treatment stifle joint was lightly packed with surgical gauze to minimize superficial contamination of tissues with contrast agent. After injection of the contrast solution was completed, the surgical gauze was removed and the joint lavaged to reduce contamination by the contrast agent on the articular and synovial surfaces of joint structures.
Contrast solution was injected through the 8F catheter placed in the caudal portion of the abdominal aorta. The injection continued until contrast solution was seen exuding from the surgical site of the treatment limb and was also observed exuding from at least 1 cut toenail/limb or was observed within small vessels (which were exposed by a skin incision) in the distal portion of the metatarsal region.
Radiographic evaluation
After the contrast agent was injected, each hind limb was radiographed to assess vascular filling. Plain film radiographsc were obtained that included the entire hind limb (proximal portion of the hind limb from the hip joint to the tarsal joint). The radiographs were subjectively assessed to ensure adequate filling of the small vessels throughout the entire limb (Figure 1). If either hind limb had poor or incomplete vessel filling, the entire cadaver (both hind limbs) was replaced in the study.
Lateral radiographic view of an entire control limb obtained from a canine cadaver that depicts the typical vascular pattern of a hind limb. Structures were as follows: 1 = femoral artery, 2 = descending genicular artery (which obliquely crosses the femur [arrow]), 3 = saphenous artery, 4 = caudal portion of the femoral artery, 5 = popliteal artery, and 6 = medial genicular artery (arrowhead). The radiographic perfusion score for this limb was 4 (scale of 0 to 4; 0 = filling of only the femoral artery, 1 = filling of the femoral artery and first arterial branches, 2 = filling of the femoral artery and secondary arterial branches, 3 = filling of the femoral artery and up to tertiary arterial branches and identification of some smaller vessels beyond the tertiary arterial branches, and 4 = uniform filling beyond the tertiary arterial branches throughout the limb).
Citation: American Journal of Veterinary Research 77, 5; 10.2460/ajvr.77.5.548
After whole limb radiographs had been obtained, the patella, patellar tendon, parapatellar tendon joint capsule, and tuberosity on the proximal portion of the tibia were harvested and radiographed by use of ultra–high-detail radiographic film.d These radiographs were evaluated to determine vascular perfusion of the harvested tissues (Figure 2).
Ultra-high-detail radiographic views of the patella, patellar tendon, parapatellar tendon joint capsule, and tuberosity on the proximal portion of the tibia for representative stifle joints subjected to MA (arthrotomy to the right [A]), LA (arthrotomy to the left [B]), MA and LA with tibial tuberosity transposition (C), and MA with TPLO (arthrotomy to the left [D]) and an unoperated control joint (E). The patella and tibial tuberosity have been darkened to increase radiographic detail. In panels A through D, there are various amounts of surface contamination of the distal region of the tendon attributable to spilled contrast material secondary to arthrotomy and small bone fragments attributable to tibial osteotomy during collection of the tissue. Bar = 10 mm.
Citation: American Journal of Veterinary Research 77, 5; 10.2460/ajvr.77.5.548
Radiographic evaluation of vascular filling and perfusion
Radiographs of the limbs obtained after injection of the contrast were reviewed by 1 investigator (MDJ) who was not aware of the surgical procedure performed on each limb. However, the investigator was aware of limbs subjected to tibial tuberosity osteotomy or TPLO. Comparison of whole limb radiographs was performed to ensure vascular filling of both hind limbs was complete and similar. Radiographs were evaluated in random order 3 times by use of a scoring system. Each radiograph was scored on a scale of 0 to 4 (0 = filling of only the femoral artery, 1 = filling of the femoral artery and first arterial branches, 2 = filling of the femoral artery and secondary arterial branches, 3 = filling of the femoral artery and up to tertiary arterial branches and identification of some smaller vessels beyond the tertiary arterial branches, and 4 = uniform filling beyond the tertiary arterial branches throughout the limb). Lateral and craniocaudal radiographic views were scored separately. Scores for each of the 3 evaluations of the lateral and craniocaudal views were added together (maximum score, 24). A grade for vascular filling was then assigned on the basis of the summed score as follows: excellent, 21 to 24; good, 16 to 20; and acceptable, 12 to 15.
Ultra–high-detail radiographs of the harvested tissues were similarly evaluated to assess perfusion of the smaller vessels. Images were digitized prior to evaluation. Digitized images were evaluated by 1 investigator (MDJ) who was not aware of the surgical procedure performed. Images were evaluated 3 times. Each radiograph was scored on the basis of a scale of 0 to 4 (0 = no vascular perfusion visible within the tissues, 1 = random perfusion of some vessels within the tissues but without discernible vascular patterns, 2 = perfusion of the vessels nearest to the patella or the tibial tuberosity, 3 = vascular perfusion similar to that for a score of 2 but with inclusion of vessels of the paratendinous region in at least 1 side of the tendon region, and 4 = perfusion similar to that for a score of 3 but that also included branching vessels from the paratendinous vessels into or overlying the tendon region). Scores for the 3 evaluations were added together (maximum score, 12). A grade for radiographic vessel perfusion was then assigned on the basis of the summed score as follows: excellent, 10 to 12; good, 7 to 9; acceptable, 4 to 6; and poor, ≤ 3.
Tissue collection and histologic examination
En bloc collection of the patella (with associated femoropatellar ligaments and joint capsule), patellar tendon and immediately adjacent joint capsule, and tibial tuberosity was performed. Care was used during harvesting to avoid direct handling of the tendon tissues. Harvested tissues were placed in 4% paraformaldehyde (the tissue fixation solution was selected to allow immunohistologic analysis as part of another study) and refrigerated at 5°C for 18 to 24 hours. Tissues were then washed 3 times in phosphate buffer solution. Tissues were embedded in paraffin, sectioned, and stained with H&E stain for histologic examination.
Transverse sections of the proximal, middle, and distal nonosseous regions of each tendon were examined. The proximal region was just distal to the tendon origin on the patella. The middle region was at the midpoint between the patellar and tibial attachment points. The distal region was just proximal to the point of attachment to the tibial tuberosity.
Tissue samples were histologically examined. Histologic vessel counts for the presence of India ink within vessel lumens in the tendinous and immediately adjacent peritendinous tissues (perfused vessels) were manually performed for the entire section of tissue on each slide. These counts were used to determine changes in blood delivery to the tendon after arthrotomy, compared with results for the control group.
Statistical analysis
Radiographic vascular perfusion scores were compared by use of χ2 analysis with a Bonferroni correction set at P < 0.01. Radiographic scores of harvested tendons for each treatment group were compared with results for the control group by use of a Student t test.
Perfused vessel counts for each treatment group were compared with results for the control group. Total counts of all regions for each tendon as well as vessel counts within each region of tendon were examined by use of the Shapiro-Wilk test for normality; these results were not normally distributed (P = 0.002). Because of the nonparametric nature of the data and the small size of the treatment groups, the total number of vessels counted within the tendon tissues were reported as median and IQR, and total vessel counts for all 3 regions of tendon and vessel counts for each region of tendon were compared by use of a Kruskal-Wallis ANOVA with Bonferroni correction for multiple comparisons. Data for control limbs were pooled for statistical analysis.
All statistical analyses were performed with statistical software.e Significance was set at values of P < 0.05, except for Bonferroni corrections, which were set at values of P < 0.01.
Results
Radiographic evaluation of vascular filling of an entire limb
Radiographic filling of vessels by contrast agent was found to be similar between the control and treatment limbs. Radiographic filling of the 24 control limbs was rated as excellent in 11 (46%), good in 10 (42%), and acceptable in 3 (13%). Of the 24 treatment limbs, radiographic filling was rated as excellent in 10 (42%), good in 11 (46%), and acceptable in 3 (13%). These values did not differ significantly.
Radiographic evaluation of vascular perfusion of harvested tissue
Radiographic perfusion of contrast agent within small vessels of the tendon and paratendinous region was compared between the treatment groups and their control limbs (Figure 3). Median radiographic scores for the MA group were 10 (IQR, 9 to 12) for the treatment limbs and 11.5 (IQR, 10 to 12) for the control limbs; these values did not differ significantly (P = 0.10). For the LA group, median score was 10 (IQR, 8 to 11) for the treatment limbs and 11 (IQR, 3 to 12) for the control limbs; these values did not differ significantly (P = 0.37). Median score for the MALA group differed significantly between the treatment (median, 1; IQR, 0 to 3) and control (median, 11; IQR, 8 to 12) limbs. Similarly, the median score for the TPLO group differed significantly between the treatment (median, 6; IQR, 6 to 7) and control (median, 12; IQR, 10 to 12) limbs. All of the control limb median scores were graded excellent. Median scores for the MA and LA groups were graded excellent. The TPLO group median score was graded acceptable, and the MALA group median score was graded poor. Radiographic scores were consistent with the histologic results for perfused vessel counts.
Box-and-whisker plots of results for evaluation of ultra-high-detail radiographs to evaluate vascular perfusion of harvested tissues for stifle joints obtained from canine cadavers and treated (gray boxes) by use of MA (MA group), LA (LA group), MA and LA with tibial tuberosity transposition (MALA group), and MA with TPLO (TPLO group); the contralateral hind limb was not treated and served as the control sample (white boxes). There were 6 limbs/treatment group. Each box represents the IQR, the horizontal line in each box represents the median, and the whiskers represent the maximum and minimum scores in each group. Each radiograph was scored 3 times by use of a scale of 0 to 4 (0 = no vascular perfusion visible within the tissues, 1 = random perfusion of some vessels within the tissues but without discernible vascular patterns, 2 = perfusion of the vessels nearest to the patella or the tibial tuberosity, 3 = vascular perfusion similar to that for a score of 2 but with inclusion of vessels of the paratendinous region in at least 1 side of the tendon region, and 4 = perfusion similar to that for a score of 3 but that also included branching vessels from the paratendinous vessels into or overlying the tendon region). Scores for the 3 evaluations were added together (maximum score, 12). *Within a treatment, median value differs significantly (P < 0.05) from that of the control group.
Citation: American Journal of Veterinary Research 77, 5; 10.2460/ajvr.77.5.548
Histologic evaluation of vascular perfusion of harvested tissues
Median total perfused vessel counts were determined (Table 1). There was no significant difference in histologic perfused vessel counts among the 4 treatment groups. Comparison of limbs in each treatment group with those in the control group revealed a significant decrease in histologic perfused vessel counts for the entire patellar tendon in the MALA and TPLO groups. A Bonferroni-type correction for repeated measurements confirmed significant (P < 0.01) differences for both groups. Median histologic perfused vessel counts in the patellar tendons for the MA and LA groups were not significantly different from those of the control limbs (P = 0.12 and P = 0.08, respectively).
Median (IQR) values for perfused vessel counts determined on histologic slides for each treatment group* and each region of the patellar tendon.†
| Median (IQR) perfused vessel count | ||||
|---|---|---|---|---|
| Treatment group* | Proximal | Middle | Distal | Total |
| Control | 69 (26–253) | 73 (16–211) | 52 (4–146) | 182 (159–261) |
| MA | 43 (6–89) | 32 (3–65) | 21 (0–52) | 99 (7–210) |
| LA | 54 (37–100) | 32 (21–60) | 28 (22–69) | 110 (84–228) |
| MALA | 13 (5–23)‡ | 11 (2–34)‡ | 2 (0–17)‡ | 31 (8–70)‡ |
| TPLO | 10 (6–27)‡ | 9 (4–26)‡ | 9 (0–28)‡ | 39 (27–65)‡ |
Vessel counts represent the total number of perfused vessels per slide.
Stifle joints were obtained from canine cadavers, and 1 hind limb of each cadaver was treated by use of MA (MA group), LA (LA group), MA and LA with tibial tuberosity transposition (MALA group), or MA with TPLO (TPLO group); the contralateral hind limb was not treated and served as the control sample. There were 6 limbs/treatment group.
The proximal region was just distal to the tendon origin on the patella. The middle region was at the midpoint between the patellar and tibial attachment points. The distal region was just proximal to the point of attachment to the tibial tuberosity.
Within a column, median value differs significantly (P < 0.01) from that of the control group.
Comparing the change in vessel counts for each treatment group with values for its own control limb revealed no significant difference in histologic perfused vessel counts for the MA and LA groups (P = 0.18 and P = 0.36, respectively). For the MA group, the median decrease in histologically perfused vessels was 41% (IQR, 2% to 91%), whereas the median decrease for the LA group was −3% (IQR, −6% to 52%). The percentage decrease in total perfused vessel counts was significantly different for both the TPLO and MALA groups. Median decrease for the MALA group was 92% (IQR, 83% to 96%), and median decrease for the TPLO group was 79% (IQR, 65% to 83%).
Perfused vessel counts did not differ significantly (P = 0.07) among the proximal, middle, and distal regions of the control patellar tendons. Histologic perfused vessel counts were significantly (P < 0.01) decreased for all regions of the patellar tendon in the TPLO and MALA groups. There was no significant difference detected in perfused vessel counts for any region of the tendon for the MA and LA groups, compared with results for the regions in the control tendons (Table 1).
Discussion
A complete anatomic description of the vascular supply and relative contributions of component vessels of the patellar tendon in dogs is lacking. In 1 study11 in which investigators used the canine stifle joint as a model for the human knee, the general pattern of blood supply to the tendon was described as containing proximal and distal anastomotic vessels with longitudinal peritendinous vessels providing the blood supply to the length of the tendon located between the patella and tibial crest. The proximal portion of the patellar tendon also has several anastomosing blood vessels arising from the infrapatellar fat pad. There is no communicating vasculature between the endosteum of the tibia and patellar tendon. Similarly, intrinsic vessels of the patella do not communicate with intrinsic vessels of the patellar tendon, but periosteal vessels commonly communicate with peritendinous vessels proximally and distally.11 Ultra–highdetail radiographs of the patellar tendon after injection of contrast agent in the study reported here revealed a vascular structure consistent with these descriptions. Branches of the descending genicular and medial genicular arteries supply the cranial portion of the joint capsule and patellar tendon.12
Previous descriptions of the blood supply to the patellar tendon of dogs indicate increased vascular density at the proximal and distal aspects, with greater vascular density in the peritendinous tissues than in the central portion of the tendon.11,13 Vessels arising at the proximal and distal aspects anastomose in the middle third of the tendon, and vascular density is lowest in the middle third.13 We did not detect a difference in median vessel counts among the 3 regions of the patellar tendon in the control limbs. It is possible that the difference was too small to detect with our group size (type II error) or that evaluation of only 3 regions of the tendon failed to yield a difference.
Decreased blood delivery in the groups involving arthrotomy and osteotomy of the proximal portion of the tibia indicated that increased surgical invasiveness of these procedures may have substantially disrupted blood delivery to the associated tissues. This was expected for the MALA group because it was designed to maximally disrupt blood delivery. There was a similar disruption for the TPLO group. Extension of the arthrotomy and elevation of the medial aspect of the periosteum was sufficient to disrupt general blood delivery to the patellar tendon. Most of the blood supply to the tibial tuberosity of dogs also comes from the periosteum on the medial aspect.14 Increased healing rates and fewer complications are reported for humans when the medial aspect of the periosteum is left intact during tibial tuberosity transposition.15 It would seem prudent to preserve the medial aspect of the tibial periosteum and limit distal extension of an MA to minimize the effect of these procedures on blood delivery to the patellar tendon. It was also unclear, on the basis of the present study with regard to the MALA and TPLO groups, whether elevation of the periosteum or the osteotomies played a greater role or whether the combination of these procedures was responsible for the loss of blood delivery to the patellar tendon in these groups. Although we were able to detect a significant difference in the presence of contrast agent for certain treatment groups, it was not possible to determine from the present cadaveric study whether this decrease would be clinically relevant. A prospective clinical study conducted to examine blood flow within the patellar tendon before surgery, immediately after surgery, and in the weeks following surgery would be needed to determine the clinical relevance of the findings for the present study.
An additional group undergoing only tibial tuberosity transposition would have been beneficial to elucidate the periosteum's contribution to the patellar tendon blood supply.12 It was also possible that the increased length of the medial incision, rather than the osteotomy, damaged the medial geniculate artery or its immediate branches to the patellar tendon. Miniarthrotomy (a limited access incision extending from the tibia to the level of the distal portion of the patella that does not involve the femoropatellar ligament) or arthroscopy procedures are less likely to damage the blood supply to the proximal portion of the patellar tendon via the routes affected in the present study, but these procedures (miniarthrotomy or arthroscopy) may still disrupt blood supply to the distal portion of the patellar tendon via disruption of the infrapatellar fat pad or when combined with osteotomy of the proximal portion of the tibia. Additional studies are necessary to determine the clinical importance of these findings.
The present study was conducted to determine whether the surgical approach had an effect on blood delivery to the patellar tendon. A potential limitation of this study was that it involved the use of cadavers. Physiologic variables of blood delivery were not reproduced for infusion of contrast agent. Additionally, heparinized solution was not infused into dogs before euthanasia, so postmortem clot formation may have affected the results. However, because both hind limbs (control and treatment) were infused simultaneously and the results compared, it was believed that this technique could be used to determine whether the surgical procedure created a difference in blood delivery to the patellar tendon. Uniformity of radiographic vascular results between the control and treatment limbs supported the validity of the findings in this study.
Most of the hypotheses regarding patellar tendinosis in human medicine involve repetitive use and cyclic fatigue or microtrauma and tendon degeneration.16–19 It has been found for the Achilles tendon in humans that blood flow is interrupted during loading of the tendon.20 During repetitive use, this may expose the tendon to periods of relative hypoxia and microtearing of the tendon.16–19 We hypothesized that surgical intervention for TPLO in canine patients may result in similar tissue hypoxia to the patellar tendon and create an environment that may contribute to tendon degeneration similar to that seen for human athletes with patellar tendinosis (ie, jumper's knee). This may be especially relevant, considering that postoperative canine patients generally bear considerably less weight on the operated limb during the first 4 weeks after surgery when patellar tendinopathy develops.
Although the underlying causes of patellar tendinosis in humans and patellar tendinopathy in dogs differ, the histopathologic changes and clinical effects appear to be similar. Patellar tendinosis in humans is a spontaneous condition seen most frequently in jumping athletes, whereas patellar tendinopathy in dogs is seen following TPLO. Both conditions include thickening of the patellar tendon and focal pain elicited during palpation over the tendon and are diagnosed on the basis of the medical history and physical examination findings, with or without ultrasonographic imaging of the tendon.1,16,21 Ultrasonography of subjects with both conditions has revealed thickening of the tendon substance with a hypoechoic region or regions within the area of discomfort.1,16,21 These conditions also have similar pathological changes. Only 2 histologic examinations of dogs with patellar tendinopathy have been reported.1 That histologic description matched the histologic criteria for patellar tendinosis in humans: collagen fiber disorganization, fibrocartilaginous metaplasia, cellular and capillary proliferation, and increased mucoid (myxomatous) ground substance. Inflammation was not a feature in either condition.1,16,21 This condition has been characterized as patellar ligament desmitis,2 but patellar tendinopathy is a more accurate characterization of the noninflammatory, degenerative nature of this condition.f
Hypovascular areas in the central portion of tendons are commonly associated with degeneration and rupture, but there is no direct evidence to prove that hypovascularity is a primary cause of tendon rupture.16,17,21 The role of the vasculature in pathological tendon conditions is often undetermined and sheds doubt on whether vascular invasion is desirable under some circumstances.22 It has been hypothesized that a decreased vascular supply to the patellar tendon insertion region may cause ischemia and subsequent degeneration of a portion of the tendon.22 Tendon degeneration may lead to reactive capillary proliferation and prominent angiogenesis seen in histologic examination of samples obtained from humans with patellar tendinosis16,22,23 and dogs with tendinopathy.1 Decreased blood delivery in the study reported here and other studies1,16,18,19,21,22 has been described for vessel proliferation in patellar tendinopathy, which indicates that measurement of blood flow and oxygen tension within the tendon of live dogs is needed to further elucidate the role that hypoxia plays in diseases and healing processes of the patellar tendon.16,22 Pain in humans with patellar tendinosis has not been associated with the presence of structural changes alone; instead, it is associated with the presence of neovascularization.16,18,19,22 Additionally, it has been found that obliteration of neovascularization and associated small nerve fibers within the tendon reduces pain associated with patellar tendinosis.24,25
In the present study, arthrotomy of the stifle joint combined with osteotomy of the proximal portion of the tibia via TPLO or tibial tuberosity transposition significantly diminished blood delivery throughout the patellar tendon in canine cadavers. For these reasons, it may be prudent, whenever feasible, to limit the extent of arthrotomy or elevation of the medial aspect of the periosteum (or both) in an attempt to preserve as much of the blood supply to the patellar tendon as possible.
Acknowledgments
Supported by the Companion Animal Health Fund at the University of Saskatchewan.
Presented in abstract form at the 2008 American College of Veterinary Surgeons Symposium, San Diego, October 2008.
ABBREVIATIONS
| IQR | Interquartile range |
| LA | Lateral arthrotomy |
| MA | Medial arthrotomy |
| TPLO | Tibial plateau leveling osteotomy |
Footnotes
Chandler JC, Egger EL, Santoni BG, et al. Effect of tibial plateau leveling on patellar tendon strain and quadriceps tension in cranial cruciate-deficient stifles: an in vitro experimental study (abstr), in Proceedings. 2nd World Congr 33rd Annu Vet Orthop Soc Meet 2006;170.
Offutt RL, Schulz KS, Walls CM, et al. The in vitro effects of tibial plateau leveling osteotomy on canine stifle extensor structures (abstr). Vet Surg 2006;35:E20.
Kodak X-Sight L/RA film, Eastman Kodak Co, Rochester, NY.
Kodak Biomax XAR film, Eastman Kodak Co, Rochester, NY.
Statistix 8, Analytical Software, Tallahassee, Fla.
Rumian AP, Wallace AL, Birch HL. P21 patellar tendon or patellar ligament? A comparative study in the ovine model (abstr). J Bone Joint Surg Br 2008;90-B(suppl II):375.
References
1. Carey K, Aiken SW, DiResta GR, et al. Radiographic and clinical changes of the patellar tendon following tibial plateau leveling osteotomy: 94 cases (2001–2003). Vet Comp Orthop Traumatol 2005; 18: 235–242.
2. Mattern KL, Berry CR, Peck JN, et al. Radiographic and ultrasonographic evaluation of the patellar ligament following tibial plateau leveling osteotomy. Vet Radiol Ultrasound 2006; 47: 185–191.
3. Pacchiana PD, Morris E, Gillings SL, et al. Surgical and postoperative complications associated with tibial plateau leveling osteotomy in dogs with cranial cruciate ligament rupture: 397 cases (1998–2001). J Am Vet Med Assoc 2003; 222: 184–193.
4. Section VII: the hindlimb. In:Piermattei DL, Johnson KA, eds. An atlas of surgical approaches to the bones and joints of the dog and cat. 4th ed. Philadelphia: Saunders, 2004; 338–355.
5. Kowaleski MP, Boudrieau RJ, Pozzi A. The stifle joint. In: Tobias KM, Johnston SA, eds. Veterinary surgery small animal. St Louis: Elsevier, 2012 980.
6. Degner DA, Walshaw R, Fowler JD, et al. Surgical approaches to recipient vessels of the head and neck for microvascular free tissue transfer in dogs. Vet Surg 2004; 33: 200–208.
7. Degner DA, Walshaw R, Fowler JD, et al. Surgical approaches to recipient vessels of the fore- and hind limbs for microvascular free tissue transfer in dogs. Vet Surg 2005; 34: 297–309.
8. Moore DC, Leblanc CW, Muller R, et al. Physiologic weightbearing increases new vessel formation during distraction osteogenesis: a micro-tomographic imaging study. J Orthop Res 2003; 21: 489–496.
9. Nonaka H, Akima M, Hatori T, et al. Microvasculature of the human cerebral white matter: arteries of the deep white matter. Neuropathology 2003; 23: 111–118.
10. Ndoye J-M, Dia A, Ndiaye A, et al. Arteriography of three models of gastric oesophagoplasty: the whole stomach, a wide gastric tube and a narrow gastric tube. Surg Radiol Anat 2006; 28: 429–437.
11. Alm A, Stromberg B. Vascular anatomy of the patellar and cruciate ligaments. Acta Chir Scand Suppl 1974; 445: 25–35.
12. Evans HE. The heart and arteries. In: Evans HE, ed. Miller's anatomy of the dog. 3rd ed. Philadelphia: Saunders, 1993; 663–666.
13. Soldado F, Reina F, Yuguero M, et al. Clinical anatomy of the arterial supply of the human patellar ligament. Surg Radiol Anat 2002; 24: 177–182.
14. Kanamiya T, Naito M, Ikari N, et al. The effect of surgical dissections on the blood flow to the tibial tubercle. J Orthop Res 2001; 19: 113–116.
15. Kanamiya T, Naito M, Hara M, et al. Tibial tubercle transfer on a medial periosteal pedicle—a report of a new technique. The Knee 2006; 13: 469–473.
16. Khan KM, Bonar F, Desmond PM, et al. Patellar tendinosis (jumper's knee): findings at histopathologic examination, US, and MR imaging. Victorian Institute of Sport Tendon Study Group. Radiology 1996; 200: 821–827.
17. Cook JL, Malliaras P, De Luca J, et al. Vascularity and pain in the patellar tendon of adult jumping athletes: a 5 month longitudinal study. Br J Sports Med 2005; 39: 458–461.
18. Cook JL, Malliaras P, De Luca J, et al. Neovascularization and pain in abnormal patellar tendons of active jumping athletes. Clin J Sports Med 2004; 14: 296–299.
19. Gisslen K, Alfredsom H. Neovascularisation and pain in jumper's knee: a prospective clinical and sonographic study in elite junior volleyball players. Br J Sports Med 2005; 39: 423–428.
20. Ohberg L, Lorentzon R, Alfredson H. Neovascularization in Achilles tendons with painful tendinosis but not in normal tendons: an ultrasonographic investigation. Knee Surg Sports Traumatol Arthrosc 2001; 9: 233–238.
21. Hamilton B, Purdam C. Patellar tendinosis as an adaptive process: a new hypothesis. Br J Sports Med 2004; 38: 758–761.
22. Fenwick SA, Hazleman BL, Riley GP. The vasculature and its role in the damaged and healing tendon. Arthritis Res 2002; 4: 252–260.
23. Yu JS, Popp JE, Kaeding CC, et al. Correlation of MR imaging and pathologic findings in athletes undergoing surgery for chronic patellar tendinitis. Am J Roentgenol 1995; 165: 115–118.
24. Ohberg L, Alfredson H. Sclerosing therapy in chronic Achilles tendon insertional pain—results of a pilot study. Knee Surg Sports Traumatol Arthrosc 2003; 11: 339–343.
25. Ohberg L, Alfredson H. Ultrasound guided sclerosis of neovessels in painful chronic Achilles tendinosis: pilot study of a new treatment. Br J Sports Med 2002; 36: 173–177.
