Keywords
hip joint - needle arthroscopy - dogs - ex vivo
Introduction
Joint disorders are commonly encountered in dogs, especially those affecting the coxofemoral joint.[1]
[2] Hip morphology can be assessed by radiography or computed tomography (CT) but evaluation of the cartilage surface is limited.[1]
[3]
[4] Arthroscopic surgery offers several advantages compared with open surgery, such as improved visualization of intra-articular structures, reduced surgical morbidity, and minimized postoperative pain.[4]
[5]
[6]
[7] In dogs, hip arthroscopy is used as a diagnostic tool for assessing cartilage lesions, aiding in the collection of representative intra-articular tissue samples, assisting in toggle rod stabilization, and contributing to fracture repair.[5]
[8]
[9]
[10] However, this technique is not without risk and iatrogenic cartilage damage is possible, especially in small animals and those suffering from osteoarthritis, where the scope can be particularly difficult to introduce.[5]
[8]
[9]
[10]
Needle arthroscopes are small-caliber portable arthroscopes of 1 to 1.9 mm in diameter and are widely used in human medicine for knee,[11] elbow,[12] and shoulder[13] exploration especially. These scopes are more flexible than traditional arthroscopes and are frequently disposable. Their use is becoming more common in veterinary medicine, offering increased feasibility and safety. To date, needle arthroscopy (NA) has been described for evaluating canine stifle,[14] canine shoulder,[15]
[16] canine elbow,[17]
[18]
[19] and feline coxofemoral joints,[20] but its application in canine hip joints has never been documented.
The purpose of this study was thus to describe and evaluate the feasibility of NA in the canine coxofemoral joint. We hypothesized that (1) there would be no difference in visualization of intra-articular structures or image quality between using NA and standard arthroscopy (SA), (2) and that there would be no difference in cartilage lesion surfaces compared with those obtained with a standard arthroscope.
Materials and Methods
Specimens
Twelve hip joints from six canine cadavers, euthanized for reasons unrelated to this study, were included. This study was approved by the VetAgro Sup Ethics and Animal Welfare Committee. A CT scan was performed on each pelvis preoperatively to assess for any joint pathology and to determine the total articular surface of each coxofemoral joint. Specimens were stored at −20°C until thawing at room temperature 24 hours prior to the study.
Arthroscopic Procedure
The six cadavers were positioned for arthroscopy in lateral recumbency, with the hip of interest uppermost. The pelvis was secured by a custom-made holding device with three pins manually inserted on a board to simultaneously apply a counter-pressure on the ischiatic tuberosity and maintain the pelvis during the procedure. Joint space distraction was achieved by applying distally directed traction to the distal femur. For each pelvis, coxofemoral SA was initially conducted on one side, and NA was then performed on the contralateral side. The SA side was randomly selected for each cadaver, to obtain an equal amount of left and right sides for each technique. Each arthroscopy was performed by a single ECVS diplomate with more than 10 years of arthroscopy experience (T.C.).
Description of the Modified Arthroscopic Approach and Instrumentation
The insertion technique was the same for both the nanoscope and the standard arthroscope. A 16-gauge catheter (Terumo Versatus Medical Corporation, New Jersey, United States) was inserted perpendicularly to the skin cranially and dorsally to the greater trochanter, with slight distal limb traction by the assistant. The joint was then aspirated and distended with 10 mL of saline solution. A flexible nitinol guidewire was subsequently inserted into the catheter (percutaneous insertion kit, Nano Arthroscopy, Arthrex, Naples, Florida, United States) allowing for catheter removal. A 3- to 5-mm stab incision was created in a proximal-to-distal direction along the nitinol guidewire using a no. 11 blade. The scope cannula was inserted over the guidewire, which was then removed, and the camera was then slid into the cannula. For SA, a 1.9-mm, high-definition, 30-degree arthroscope (Karl Storz SE & Co. KG, Tuttlingen, Germany) was used with a 2.5-mm outer diameter cannula. For NA, the same 1.9 mm, 0-degree arthroscope (NanoScope, Arthrex Inc., Naples, Florida, United States) was used for each hip with a 2.2-mm outer diameter sheath.[21] Fluid egress was achieved with an 18-gauge needle inserted craniolaterally into the joint, and a constant 40 mm Hg flow of saline solution was provided by a motorized pump (Arthrex Continuous Wave III, Arthrex Vet Systems, Naples, Florida, United States). Moderate distal limb traction was applied throughout the procedure.
Intraoperative Assessments
For each procedure, the difficulty of scope introduction was quantified (0 = easy, achievable on the first attempt; 1 = fewer than 3 attempts required; 2 = more than 3 attempts required; 3 = not feasible). Arthroscopic hip evaluation was performed using a systematic technique to identify intra-articular structures and all identifiable intra-articular structures of interest were documented. A score out of 6 corresponding to the identification of the 6 main structures (femoral head, acetabulum, labrum, synovium, transverse ligament, and ligament of the head of the femur) was also measured (1 = visualized; 0 = not visualized). Identification of iatrogenic cartilaginous lesions during surgery was also noted, and their precise locations were documented. A prediction of these lesions was also provided based on the surgeon's (T.C.) subjective feeling during scope insertion.
Gross Dissection
Postoperative dissection of the coxofemoral joint through a medial approach with incision of the transverse ligament allowed for the identification of cartilaginous lesions. Location, size (width, length or diameter, and shape), and thickness (partial thickness when no subchondral bone was exposed; full thickness when subchondral bone was exposed) of these lesions were detailed after the application of 0.5 mL of India Ink (Winsor and Newton Ink, London, England) followed by thorough rinsing with sterile saline solution 1 minute after application. The cartilage was photographed perpendicular to the joint surface and the digital images were saved ([Fig. 1]).
Fig. 1 Photograph of two femoral heads after dissection, section of the round ligament, and application of India ink (0.5 mL) (Winsor and Newton Ink, London, England) followed by rinsing with sterile saline. Iatrogenic articular cartilage injuries (ACIs) are colored black and highlighted by the white arrows. A needle arthroscopy (NA) was performed in (A) and a standard arthroscopy (SA) was performed in (B).
Calculation of Injured Articular Surfaces
After dissection, ACI were measured after India ink staining. For this purpose, a Castroviejo caliper (VAR 4000 CAL, Arthrex Vet Systems) was used, and lesions were categorized as linear or punctate. For linear lesions, measurements of width (w) and length (l) were taken, and for punctate lesions, the diameter (r) was measured. The ACI was calculated as the sum of the surfaces of each lesion for each joint (with a rectangular surface of l * w; for linear lesions and a circular surface of π * r
2 for punctate lesions). ACI were classified as minor and non-significant when the surface represented less than 10% of the total cartilage area (TCA).[22]
Blind Comparison of Visual Qualities between the Two Arthroscopic Methods
For each targeted intra-articular structure listed above, four images were extracted from each procedure (NA and SA). Images showing a similar structure for each procedure were randomly matched. Three independent surgeons (one ECVS diplomate, a board-eligible surgeon, and an ECVS resident) were asked to judge the quality of images, by subjectively choosing the best image between the two appearing on the screen. Since the needle arthroscope images are round and those of the standard arthroscope are square, the SA group images were cropped in a circle using photo editing software (Adobe Photoshop Lightroom, Version 7.4.1), to make them uniform and indistinguishable ([Fig. 2]).
Fig. 2 Example of image processing to standardize arthroscopic views of the needle arthroscope (left) with those of the standard arthroscope (right). These pairs of images were then displayed to the observers, who had to choose the one they thought was of better quality. Note the larger field of view offered by the needle arthroscope.
Statistical Analysis
Sensitivity and specificity of SA and NA examination were measured for the prediction of cartilage lesions occurrence during arthroscopy.
The normality of the data was tested using a Shapiro–Wilk test and quantile–quantile plots. A nonparametric model was used, with values expressed in median (first quantile–third quantile). The number of structures identified, and the surface area of cartilage lesions were compared according to the arthroscopic methods used, using Kruskal–Wallis tests. The photographic quality of one of the two methods, as well as the differences between the blind observers, were tested using a chi-square test. A p-value of less than 0.05 was considered significant.
Results
Signalment and Preoperative Computed Tomography Scan Results and Measurements
Twelve coxofemoral joints from six canine cadavers were included in this cadaveric study. All dogs were 5-month-old Beagles with a postmortem median body weight of 9.6 kg (7.8–12.4). No joint pathology was detected on CT images and the median total articular surface was calculated as 443.69 mm2 (403.13–530.67) for the acetabulum and 511.77 mm2 (479.69–543.14) for the femoral head.
Surgical Technique
Six SAs and six NAs were performed with an equal distribution of right and left sides. First, approach attempts were successful in all six joints for SA and in all but one joint for NA, where less than three attempts were required (level score of 1 with two attempts needed). Identification of all six targeted intra-articular structures was possible in all 12 joints (score of 6/6) ([Fig. 2]). For one NA case, a difficulty was reported in visualizing the transverse ligament. For one SA case, a difficulty was reported in visualizing the labrum. Articular cartilage surface was observed in all joints and was grossly normal in five joints. Of the seven joints (5 NA and 2 SA) with ACI observed during arthroscopy, a small, linear, partial thickness lesion was identified on the femoral head in seven cases at the caudoventral region (insertion site of the scope), and on the caudal portion of the labrum in one case ([Fig. 3]). During arthroscopic approach, the surgeon predicted cartilage damage in seven cases (5 NA and 2 SA) due to rubbing of the insertion needle or scope. Of these seven cases, only five (4 NA and 1 SA) showed ACI per-arthroscopically.
Fig. 3 Needle arthroscopic aspect of the cartilage surface of the femoral head (FH) and acetabular (AC), the transverse ligament (TL), the ligament of the head of the femur (LHF), the labrum (La), and the synovium (SY). Numerous structures can be seen on the same picture simultaneously. Articular cartilage injuries (ACI) are visualized on the last picture (black arrowheads).
Gross Dissection and Calculation of Injured Articular Surfaces
Articular cartilage injury (ACI) was documented in all cases in the NA group and in all but one case in the SA group. Iatrogenic cartilaginous lesions were suspected per-arthroscopically in 5 NA cases and 2 SA cases, showing that 83% of lesions caused by NA were anticipated, compared with 40% for SA. On the femoral head surface at the site of scope insertion, minor partial thickness ACI consisting of one to three linear abrasions of 1 to 3 mm length were observed in 10 hips (5 NA and 5 SA), and a focal full thickness was noted in two hips (1 NA and 1 SA). On the acetabular surface at the site of scope insertion, only three cases (2 NA and 1 SA) showed minor partial thickness ACI. These ACI were graded minor (<10% TCA) in all cases. The median surface of ACI represented 0.09% (0.06–0.21) of the total cartilaginous surface in the NA group and 0.11% (0.02–0.33) in the SA group.
Blind Comparison of Visual Qualities between the Two Arthroscopic Methods
Of the 24 comparisons submitted, NA images were judged superior in quality in 100% of cases for the first surgeon, 95% of cases for the second one and 86% of cases for the third one.
Statistical Analysis
Concerning the prediction of articular cartilage lesion by SA and NA examination, the sensitivity and specificity were 0.56 and 0.33, respectively. No statistical difference was found in the median surface of ACI between NA and SA methods (p = 0.94). During the blind comparison of visual qualities between NA and SA methods, there was no statistical difference between the three observers (p = 0.15) and NA images were statistically better than SA images (p < 0.001).
Discussion
The main findings of this study were that needle arthroscopic evaluation of the canine hip was successfully performed in all hips and was not associated with significant damage to intra-articular structures. Comparison with 1.9-mm hip SA showed that NA was superior in terms of image quality and that there was no significant difference for ACIs. To the author's best knowledge, this is the first report of NA in the canine coxofemoral joint.
An arthroscopic hip exploration technique similar to that previously described was performed in this study with the additional use of an insertion kit with a flexible guidewire for both the NA and SA groups.[5]
[6] These supplementary steps in the procedure gave satisfactory results with a safer and more reliable scope insertion into the congruent hip joint. Scope insertion was subjectively found by the operating surgeon to be easy to perform in all hips. In all SA cases and in all but one case of the NA group, the scope could be inserted on the first attempt. In this particular NA case, the scope was able to be inserted on the second attempt. The scope cannula was inserted at the 12 o'clock position relative to the femoral head, as previously established, and the fluid egress was cranial to the scope portal.[5]
[6] NA of the canine hip was feasible and subjectively straightforward in cadaver hips.
Joint exploration was performed successfully in all cases using an arthroscope and an egress needle only. Intra-articular structures of interest were visualized in all hips within both groups. The quality of visualization of structures evaluated in the NA group was significantly superior to the intra-articular structures assessed in the SA group based on the blinded evaluation of three independent surgeons. The images from the NA group were perceived to be sharper, and it was easier to distinguish between the different structures in the NA group. This result can be explained by the greater field of vision provided by the Nanoscope (120 degrees) compared with the standard 1.9-mm arthroscope, with a field of vision ranging from 80 to 90 degrees. Furthermore, the 0-degree angle view of the needle arthroscope was not a limiting factor for joint exploration in this study. On the contrary, evaluation of the labrum and the synovium was perceived to be easier and more complete in the NA group compared with the SA group. To the best of our knowledge, there is no comparison in the veterinary literature between the image quality of a Nanoscope and a standard arthroscope of the same size. The available data currently compared a Nanoscope with a larger standard arthroscope, which offers superior image quality due to its wider field of view, stronger lighting, and greater irrigation flow through the cannula.[14]
[18] Interestingly, the quality of the images did not seem to be affected by the reuse of the Nanoscope. These devices are usually intended for single use, but this was not a drawback in this study. Furthermore, the evaluation of the quality of images was not impacted by the surgeon's experience as no significant difference was found between the three observers. The Nanoscope therefore appears to be an accessible, high-quality instrument.
A small degree of ACI occurred in all joints, with only a median of 0.09 and 0.11% of the total cartilaginous surface affected in the NA and SA groups, respectively. ACI lesions were therefore all considered minor. Those results are similar to previous studies, most notably those evaluating SA in the canine stifle, and NA in the feline hip joint.[20]
[22]
[23]
[24] ACI can occur during scope insertion and manipulation, portal establishment, or by the needle egress. In our study, most of the lesions were located in the craniodorsal region of the femoral head, where the scope was inserted. Few lesions were present on the acetabular cartilage surface. The majority of lesions were partial thickness ACIs. Iatrogenic cartilage injuries are a common complication associated with arthroscopy and their consequences are currently uncertain.[23] Our results showed that both needle and SA are not without risk. Techniques to reduce iatrogenic lesions must be found and implemented, as experimental studies have shown that full-thickness cartilage lesions in dogs can induce progressive osteoarthritis.[23] One of the strategies found is joint distraction.[25]
[26] Although one could assume that the use of small diameter arthroscopes would limit ACI (and that the needle arthroscope would therefore be of interest), this assumption could not be verified in this study, because the nanoscope's diameter was similar to that of the 1.9-mm arthroscope.
Nanoscopy shows undeniable advantages, such as being rapid and easy for surgeons to set up, which means that it can be used more widely.[15]
[18] In addition, the increased length of the sheath compared with a standard 1.9-mm arthroscope may be of interest for large dogs with a large amount of muscle mass (gluteal muscle) to pass through to insert the scope. Further research could determine whether NA would be responsible for fewer ACI lesions in these large dogs that require a longer (and therefore larger diameter) standard arthroscope. One of the possible disadvantages with NA is the irrigation flow rate, which can be slower than with SA, although this was not a limiting factor for the use of an irrigation pump in this study. In addition, the length of the sheath for small animals can be more challenging, with an increased distance from the surgeon's hand and thus excessive mobility.
Limitations of this study include the ex vivo nature of the study with a relatively small number of juvenile dog's cadavers without hip pathology. Care should be taken when interpretating these results in other conditions such as osteoarthritis. Insertion of the scopes would probably have been more complicated, and more lesions would have been visualized, such as synovitis or noniatrogenic cartilaginous lesions. However, the absence of cartilaginous lesions on young hips allowed the quantification of iatrogenic lesions in this study. Furthermore, postmortem alterations can compromise articular cartilage, making it more susceptible to ACI.[20]
[27] Although the quality of visualization of structures between the two groups was compared subjectively by three independent surgeons, its interpretation should be used with caution. Finally, the 1.9 mm diameter was the only size used, so our results may not apply directly to the results obtained by using a larger diameter traditional arthroscope for hip arthroscopy in dogs.
Conclusion
This is the first study reporting the use of a 1.9-mm diameter, 0-degree needle arthroscope for exploration of the hip joint in dogs. We found that NA of the canine coxofemoral joint is a safe and feasible minimally invasive procedure, which provides better image quality than a standard 1.9-mm, 30-degree arthroscope, without significant difference in the amount of iatrogenic cartilage lesions. NA may be considered as an option for the exploration of the canine hip joint. These preliminary results, on healthy and juvenile hips, are the first step toward future clinical studies on pathological hip joints.
