Plasma-Mediated Bipolar Radiofrequency Ablation of Overlong Soft Palate in the Dog: A Pilot Study
ABSTRACT
The objective of this study was to compare the clinical, biological, macroscopic, and histologic outcomes after resection of the soft palate by plasma-mediated bipolar radiofrequency ablation (PBRA) or traditional incisional techniques (incisional soft palate resection [INC]) in dogs.
Ten dogs were divided in two groups. In the INC group, the soft palate was incised with scissors and the wound was sutured in a continuous pattern. In the PBRA group, a wand was used to ablate the desired portion of the soft palate, without suture. Clinical, biological, macroscopic, and histologic assessments were scheduled over 14 days.
The duration of surgery was significantly shorter for the PBRA group. The C-reactive protein concentrations were significantly higher in the PBRA group at 6 hr and on day 3 (P < .05) but with values very close to the baseline. C-reactive protein concentrations were maximal, but with low values (<25 mg/L), at day 1 for both techniques. The irregularity scores for the soft palate caudal border on days 1, 3, and 14 were significantly higher in the INC group than in the PBRA group (P < .05). The main histopathologic changes were the presence of superficial granulomas and a significantly greater depth of tissue damage in the INC group (2.5 ± 0.3 mm) compared with the PBRA group (1.5 ± 0.1 mm; P < .05).
PBRA compared favorably with the traditional technique in terms of ease, duration of surgery, and depth of tissue damage. Future studies are warranted to validate its effectiveness for treating brachycephalic airway obstruction syndrome in dogs.
Introduction
Elongated soft palate is a frequent congenital respiratory problem in brachycephalic dogs. The condition is usually associated with other abnormalities of the upper respiratory tract such as stenotic nares and everted laryngeal saccules.1 The free border of the soft palate overlaps the glottal opening, causing increased respiratory effort, inspiratory distress, inflammation, and edema. The condition is self-worsening, warranting early surgical correction.1
Traditionally, soft palate resection is performed by sharp incision with a scalpel blade or scissors followed by apposition and suture of the oral and nasal mucosal ends.2 Alternative techniques of soft palate resection such as CO2 laser, bipolar sealing device, and harmonic scalpel have been introduced to reduce postoperative inflammation and for their superior ease of use.2–6 An alternative technique, folded flap palatoplasty, has been developed to correct both the excessive length and thickness of the soft palate and to relieve nasopharyngeal obstruction.7,8 This technique produces less postoperative pharyngeal inflammation and edema because the suture material is more rostrally situated. However, this technique is slightly more invasive and more technically demanding than conventional soft palate resection.7,8 Recently, objective clinical effects of the folded flap technique associated with modified rhinoplasty, ventriculectomy, and partial cuneiformectomy have been assessed by whole-body barometric plethysmography, with encouraging results.9
Plasma-mediated bipolar radiofrequency ablationa (PBRA) is a recent radiofrequency-based surgical modality commonly used in human surgery.10 With this technology, radiofrequency energy is applied to a conductive medium (usually saline solution), which causes a highly focused plasma field to form around the energized electrodes. This plasma field is composed of highly ionized particles, which have sufficient energy to break the organic molecular bonds within tissue. The byproducts of this non–heat-driven process are elementary molecules and low-molecular-weight inert gases. PBRA causes volumetric removal of tissue at low temperatures ranging from 60°C to 80°C, with minimal collateral damage.10 The histologic effect of PBRA was examined in several studies.11–20 Most of these were histopathologic evaluations of human tissue resected by PBRA, and included tonsils, nasal turbinates, and ovarian cysts.11,16–18 Two others were cadaveric studies of sheep and pig intervertebral discs.12,15 Four studies described the serial appearance of the wound created with PBRA following superficial application of the probe and a follow-up period of 3, 7, 14, and 29 days.13,14,19,20 The examined tissues included canine vocal fold, rat tongue, sheep paranasal sinuses, and rabbit paranasal sinuses.13,14,19,20 Even though PBRA is largely used in human medicine for surgery to reduce snoring, no scientific studies have shown its tissue effects over a period of time.21 To our knowledge, no assessment of the macroscopic and histologic appearance of the soft palate after resection of its free border with PBRA technology has been done in any species.
The purpose of our study was to examine the serial appearance of the wound over 14 days after resection of the soft palate in healthy nonbrachycephalic dogs and to compare the clinical, biological, macroscopic, and histologic outcomes of PBRA, used in cutting mode, with the traditional incisional technique. Our hypothesis was that PBRA-assisted palatoplasty would be faster and easier than the traditional technique and would result in less postoperative edema and inflammation.
Materials and Methods
Dogs
Ten adult beagle dogs (10.3 ± 1.6 kg; range 9–12.5 kg) were randomly divided into two groups of five dogs for either PBRA soft palate resection or incisional soft palate resection (INC). The dogs were acclimated to the new accommodation room and to being manipulated by examiners for 7 days prior to the study. Each dog underwent a complete physical examination and serum biochemical analyses prior to surgery. Before being enrolled in our study, these beagle dogs had already been used as control dogs in a previous procedure and were intended to be euthanized. Their general state of health and well-being had been fully restored before the current study. Given that the current study was classified as a nonrecovery procedure, it was conducted in total agreement with the Directive 2010/63/EU of the European Parliament and of the Council on the protection of animals used for scientific purposes. The study was also approved by the Institutional Ethical Committee.
Surgical Technique
All dogs in both groups were induced with propofol (6 mg/kg intravenously) and received morphine (0.2 mg/kg intravenously). The dogs were intubated and maintained with isoflurane (2%) in oxygen.
They were then positioned in sternal recumbency with the head raised by means of a tape placed under the maxilla. A 5 mm-wide segment of the caudal border of the soft palate was resected in both groups. The two procedures were performed by the same surgeon (S.P.). A PBRA system consisting of a consoleb and wandc attached to a line of saline for automatic continuous irrigation was used for the PBRA procedure. A new wand was used for each dog. The cutting mode was selected with a power setting of 7. The tip of the soft palate was grasped with an Allis forceps, and the wand was used to ablate the desired portion of soft palate. No suture was placed. For the INC procedure, an incision was made 5 mm from the free edge of the soft palate with Metzenbaum scissors. A muco-mucous suture (polydioxanone, 4/0–dec 1.5 on swaged round needles)d was applied in a continuous pattern.
All dogs were allowed to recover, then extubated and monitored hourly for clinical signs associated with the respiratory tract, over a total of 6 hr. Surgery duration and any incidental event were noted for both procedures. Morphine (0.1 mg/kg subcutaneously) was given for postoperative analgesia 4 hr after the first injection. Additional morphine was reinjected every 4 hr, depending on the pain score (see below).
Postoperative Evaluation
Clinical assessments were scheduled daily over 14 days. Biological and macroscopic assessments were performed at 1, 3, 7, and 14 days and histologic evaluation only at 14 days. Each clinical and histopathologic evaluation parameter was graded as follows: 0: normal; 1: mild changes; 2: moderate changes; 3: severe changes.
A single examiner, blinded for the surgical technique, performed a thorough physical examination before surgery, 6 hr postoperatively, and then daily from day 1 to day 14. Six parameters were used to assess pain (subjective global pain evaluation, general attitude, interactive behavior, cardiac frequency, ability to palpate the larynx, and response of the larynx to palpation). The intensity of respiratory noises was noted on auscultation. The presence of cough and gagging was also noted.
Serum samples were taken once on the day of surgery (baseline) and then 6 hr, 1 day, 3, 7, and 14 days after surgery. The C-reactive protein (CRP) concentration (reference interval: 0–10 mg/L) was measured by immunoturbidimetrye method. CRP concentrations were plotted over time, and the area under the curve was calculated for each dog.
The soft palate was inspected immediately after the surgical procedure and then, under general anesthesia with propofol (6 mg/kg intravenously), on days 1, 3, 7, and 14. A digital photograph was taken, and a score (0 = no change; 1 = mild changes; 2 = moderate changes; 3 = intense changes; 4 = severe changes) was assigned by two independent observers at the end of the study. For this macroscopic evaluation, five parameters were assessed and graded: edema, regularity of the section, thickness, color of the soft palate border, and size of the granuloma if present.
The dogs were euthanized 14 days after surgery. The remaining portion of the soft palate was removed, sutured to cardboard, and fixed in 10% buffered formalin. Five full-thickness sections (4 mm wide) were made perpendicular to the edge of the excision. The first section was obtained at the midline of the palate. Two sections were made at the left and right lateral edges and two others between the midline and the lateral edges. The sections were embedded in paraffin, sectioned at 3 μm, and stained with hematoxylin and eosin. Each palate was subjectively graded for five parameters (inflammation, fibroplasia, edema, hemorrhage, and necrosis) with the same scale used by Brdecka.22 A consensual assessment was obtained by two pathologists (M.D. and M.O.S.). The maximum depth of injury was also recorded for each sample. The highest value of the different sections was recorded for each parameter.
Data Analysis and Statistical Evaluation
An analysis of variance test for clinical, macroscopic, and histologic data was employed for the statistical analysis. A Cohen's Kappa coefficient was used to measure inter-observer agreement. The means of the observers’ values were used if this coefficient was ≥0.7. The significance level was set at P < .05. Results were expressed as mean ± standard deviation.
Results
The physical examination and serum biochemical analyses were unremarkable for all dogs at the beginning of the study. No intraoperative complications occurred in either of the groups. The duration of surgery was significantly shorter for the PBRA group (504 ± 168 s) than for the INC group (1,044 ± 202 s; P < .05).
The inter-observer agreement for all five macroscopic parameters was good, ranging from 0.7 (edema) to 0.9 (color of soft palate border). The physical examinations were unremarkable for both groups throughout the study. The pain and respiratory scores did not differ significantly between the two groups and did not vary significantly before or after surgery. No additional morphine was required for any of the dogs.
The C-reactive protein results are presented in Figure 1. In the PBRA group, CRP was high on postoperative days 1, 2, and 3. In the INC group, CRP was only increased on postoperative day 1. CRP was significantly higher for the PBRA group than for the INC group at 6 hr and on day 3 (P < .05). The area under the curve for the PBRA group (123.8 ± 14.7 mg × day/L) was not significantly different (P = .11) from that of the INC group (104.1 ± 10.6 mg × day/L).
Citation: Journal of the American Animal Hospital Association 54, 5; 10.5326/JAAHA-MS-6668
The macroscopic appearance of the soft palate of one representative dog from each group is shown in Figure 2. The mean scores for each of the five dogs and the two observers (10 observations) are presented in Table 1 for each group.
Citation: Journal of the American Animal Hospital Association 54, 5; 10.5326/JAAHA-MS-6668
In the INC group, swelling of the palate was visible on day 0, 1, and 3, with maximum edema on day 1 (scores from 0 to 2). In the PBRA group, mild edema was observed in only one dog on day 3 and by only one observer. The edema did not differ significantly between the two groups.
The soft palate was mildly to highly irregular in all dogs in the INC group from day 0 to day 14, with a peak on day 3 (scores from 2 to 4). A mildly to moderately irregular soft palate wound was also seen in all dogs in the PBRA group from day 1 to day 3 and in only one dog on day 14. In the PBRA group, severe irregularity was observed in only one dog on day 0 and its soft palate was perfectly regular by day 14. The irregularity score was significantly higher for the INC group than for the PBRA group on days 1, 3, and 14 (P < .05).
Mild to severe thickening of the soft palate was detected in all dogs in the INC group from day 1 to day 7, with a peak on days 1 and 3 (scores from 2 to 4). Mild to intense thickening of the soft palate was apparent on day 1, and moderate to severe thickening was noted on day 3 in all dogs in the PBRA group, which strongly improved by days 7 and 14 (only one dog showed mild thickening on day 14). The thickness score at the end of the surgical procedure, and on day 14, was significantly higher for the INC group than for the PBRA group (P < .05).
Two small granulomas were observed in one dog in the INC group from day 1 to day 3 in focal areas where apposition of the mucosa had been inadequate.
A discolored fringe on the free border of the soft palate, delineated by a thin reddish rim, was observed in all dogs in the PBRA group, from day 1 to day 7 (Figure 2). The extent and intensity of the discoloration diminished over time in all dogs, and the color difference was no longer apparent on day 14 (Figure 2). Whatever the changes, the color scores did not differ significantly between the two groups.
The histologic scores are listed in Table 2. Microphotographs of soft palate samples are presented in Figure 3. By day 14, the surface of the section obtained after PBRA had epithelialized even though no sutures had been used to appose the oral and nasal mucosa. Most of the histopathologic lesions observed with both methods were similar. Changes were apparent in two layers of the connective tissue. The superficial layer was thin with a moderate perivascular infiltrate of mononuclear cells (macrophages, lymphocytes, plasmocytes) and mild edema. The deep layer showed local extensive, moderate-to-marked fibroplasia and some areas of myofiber degeneration and focal necrosis. The main differences in histopathologic changes between the two methods were the presence of superficial granulomas and/or suppuration in the INC group and a significantly greater depth of tissue injury (2.5 ± 0.3 mm) in the INC group compared with PBRA group (1.5 ± 0.1 mm; P < .05).
Citation: Journal of the American Animal Hospital Association 54, 5; 10.5326/JAAHA-MS-6668
Discussion
The results of our study confirmed our hypothesis. Resection of the soft palate with PBRA was easier and faster than with the traditional incisional technique. To our knowledge, this is the first in vivo report to compare the serial development of the wounds obtained after resection of the soft palate with the PBRA technique or traditional incisional technique, over 14 days, and to assess any clinical, biological, macroscopic, and histologic differences between the techniques.
Current evidence suggests that no major safety concerns are associated with radiofrequency ablation of the soft palate for snoring in humans.23 Radiofrequency ablation can be used interstitially or in cutting mode. In human medicine, the most recent interventional procedure guidance recommends interstitial radiofrequency ablation.23 The aim with this procedure is to stiffen the soft palate in order to prevent excessive vibration but not to ablate a volume of soft palate as in our study. A wand with needle tip is directed upward toward the soft palate and makes a series of very shallow punctures in the underlying muscle. A lower energy setting is used than in our study, just to scar and tighten the tissue. It is impossible to extrapolate such results to PBRA soft palate resection by cutting mode in dogs. In humans, PBRA is used in cutting mode for tonsillectomy, with the same wand as in our study. The risk of posttonsillectomy hemorrhage was recently compared in 15,734 patients following use of the cold steel technique (gold standard) or hot techniques (diathermy scissors, PBRA, laser, and ultracision).24 PBRA resulted in a 3.2-times higher risk of posttonsillectomy hemorrhage than the cold technique but was associated with a lower risk of need for further surgery than with other hot techniques. These results therefore warranted a study of local macroscopic effects on soft palate after ablation by PBRA, over a 14-day period, as described here.
In previous studies of PBRA involving histologic assessment of the remaining tissues, PBRA was applied locally for just a few seconds on vocal fold, tongue, sinus, or turbinate mucosa without any tissue resection.13,14,19,20 As the probe was used briefly and superficially in these studies, they could not provide information about the soft tissue damage occurring in cutting mode during tissue ablation. The sizes of the tissue lesions generated by interstitial radiofrequency at differing power settings were recently determined in human tonsils and chicken breast tissue ex vivo.21 The most important reported limitation of these previous studies is that in vitro or ex vivo findings cannot necessarily be extrapolated to in vivo conditions. Furthermore, the tissue effect of PBRA can evolve over time or depend on the type of tissue, hence the need to assess the effects of PBRA, used in cutting mode on canine soft palate tissue, in vivo over a period of time corresponding to the healing period.
The present study of soft palate resection was conducted on healthy, nonbrachycephalic dogs. This was considered appropriate because our objective was to compare the serial macroscopic appearance of the wound after PBRA or use of the incisional technique, before applying the procedure in clinical patients. By using clinically normal dogs, it was possible to eliminate any variables attributable to brachycephalic syndrome abnormalities (stenotic nares, everted laryngeal saccules, or tracheal abnormalities) or specific histopathologic findings (greater thickening of the soft palate, peculiar histologic features), and also to harvest histopathologic samples at the end of the study.25–27 Any postoperative clinical and histopathologic signs would thus be directly attributable to surgery.
No adverse reaction, such as excessive inflammation, hemorrhage, edema, aberrant scar formation or their manifestation as cough, gagging, and pain, was observed during the postoperative clinical assessment, whatever the technique used.
The severity of inflammation generated by surgical trauma was assessed by measuring CRP concentrations over time. CRP was mildly increased with both techniques. The maximum CRP concentrations were obtained on day 1 with both techniques, as previously described.28,29 Although CRP was significantly higher with PBRA at 6 hr and on day 3, the area under the curve from baseline to day 14 did not differ significantly between the two groups. In recent studies, CRP varied from 0.07 to 16.8 mg/L in clinically healthy dogs and increased 17- to 29-fold from the baseline value, 24–48 hr post surgery in dogs after ovariohysterectomy and excision of superficial tumors, respectively.28,29 In the present study, the maximum CRP concentrations were low (<30 mg/L), indicating that the tissue trauma due to surgery was only slight.
PBRA palatoplasty was easy to perform in the current study and allowed precise sculpting of the soft palate, as shown by the significant differences in the regularity scores on day 1, day 3, and day 14. Resection with scissors and sutures requires soft palate manipulation and, due to its dependence on tissue elasticity and the accessibility of the area, may not be easy. With PBRA, soft palate manipulation is minimal, and the wand is ergonomic. The significantly lower scores obtained for soft palate thickness on day 0 and day 14 in the group receiving PBRA were mainly due to (1) its limited depth of effect, (2) its low temperature, and (3) the lack of suture of the soft palate free border after resection.14 As in humans, a PBRA channeling procedure could be used to achieve soft palate thinning and shrink the tissues surrounding the treatment zone after elongated soft palate ablation.30 This procedure could provide a minimally invasive alternative to the folded flap technique described to reduce soft palate thickness in dogs.7,8 An assessment of PBRA soft palate thinning was beyond the scope of our study and still needs to be explored in dogs.
A color change, visible as a white fringe on the free border of the soft palate, was observed with PBRA (Figure 2). This color change evolved favorably, with a maximum intensity on day 1 and progressive resolution until day 14. The precise nature of this modification remains unknown and warrants further investigation in histologic studies. It could be attributed to tissue ischemia, very superficial mucosal erosion, or coagulation necrosis, which was the most consistent finding observed shortly after applying PBRA.14,24 Whatever the surface area of this white fringe, it was not correlated with tissue damage severity, as shown by the lack of mucosal debridement and the limited depth of histologic injury on day 14. This effect seems to be very superficial and quickly disappears. These results are consistent with several previous studies that compared the outcome of PBRA versus laser for tonsillectomy or palatoplasty in humans.16,31
In contrast to CO2 laser, PBRA removed tissue at much lower temperatures, resulting in less peripheral thermal damage.31 Results indicated that PBRA produced less postoperative pain and improved healing. These advantages were attributed to the limited depth of effect and low temperature of PBRA (60°C–80°C). In contrast, laser treatment was considered to result in more peripheral damage with charred tissue and a much higher temperature (about 400°C).16,31 Even though PBRA is associated with posttonsillectomy hemorrhage in humans, there is less need for further surgery than for all the other hot techniques, probably because of the limited peripheral effect of PBRA.24
The local macroscopic evaluation revealed inflammatory changes on the soft palate with both techniques. Although the difference was not significant, PBRA was associated with less edema and fewer signs of inflammation, such as granulomas. At day 14, the good thickness and regularity scores obtained for the PBRA group clearly demonstrated the advantage of PBRA over the incisional technique for soft palate resection and highlighted the low morbidity of the technique, as reported in another preclinical study of PBRA on canine vocal cords.14
In earlier studies, histologic assessments of canine soft palate had been done on days 2, 3, 4, 7, and 14 after resection, either on small biopsies or after sacrifice.3,4,14 Histologic changes were reported with both laser and traditional techniques, with peak inflammation, necrosis, and ulceration occurring on day 3.4 PBRA induced initial stages of inflammation on the canine vocal fold on day 4, and a more mature inflammatory reaction was observed on day 7.14 In our study, a single histopathologic assessment was done on day 14 when the presence or absence of aberrant scar formation was most visible. The lack of histopathologic information prior to day 14 was mitigated by evaluating a systemic inflammatory response based on CRP, as in a previous study, and by local macroscopic assessment of the surgical wound.32 Nevertheless, a histopathologic examination of the earlier stages could help us better understand the injury mechanisms associated with PBRA in cutting mode. The grading system used for the histologic evaluation has already been described.22 It did not reveal any difference between the two techniques used here with regard to inflammation, fibroplasia, edema, hemorrhage, and necrosis. This might be because the histologic samples were obtained during the late phase of inflammation.4 In a study involving incisional or CO2 laser techniques, the necrosis, ulceration, and inflammation noted on days 0–3 had disappeared by day 14, and the lesions noted on day 14 were similar to those reported in our study.4 Our scores compared favorably with those of other studies involving CO2 laser or a bipolar sealing device, probably because our evaluation was done much later (on day 14 versus day 2 or day 4).22
Histologic examination revealed a significantly smaller depth of effect with PBRA than with the traditional technique. This limited depth of injury observed with PBRA was consistent with previous studies in which PBRA only affected the superficial tissue with no underlying injury.14,21 The depth of injury produced by PBRA in cutting mode in our study (1.5 ± 0.1 mm) compared favorably with that of laser treatment (3.3 mm) and a bipolar sealing device (3.5 mm), although the timing of our analysis was later than those of other studies.22
A secondary significant difference was the presence of granulomas and suppuration in all dogs in the incisional group compared with only one in the PBRA group. This beneficial effect of PBRA is probably related to the absence of a foreign body reaction due to suture materials. The choice of the suture material would have an influence on the inflammatory effects of the suture. A 4/0 monofilament absorbable suture on swaged round needle was chosen to limit tissue trauma. In a clinical situation, it would probably be preferable to choose a more rapidly absorbed suture such as polyglecaprone 25 rather than polydioxanone to reduce tissue inflammation. However, considering the time necessary for polyglecaprone 25 absorption, we assumed that at 14 days this suture would also be persistent in tissue, which would not have changed our histological findings.33
Complete healing was obtained in all dogs in the PBRA group without any need to appose the oral and nasal mucosa with sutures and without aberrant tissue formation.
The duration of surgery with PBRA (504 s) was halved in comparison with the standard technique. This reduction in operative time was also observed with laser (174–309 s), bipolar sealing device (67 s), and harmonic scalpel (300 s).4,6,22 The operative time required with PBRA should improve further as experience is gained with this new procedure.
The PBRA handpiece costs approximately $150 and, as a single-use instrument, is quite expensive for general use. However, after cleaning and sterilization with gas or liquid sterilizing agents, the wand may be reused more than 10 times for a range of soft tissue procedures (personal data). This represents a significant saving in cost and makes PBRA a financially viable alternative to sharp dissection or lasers.
The first limitation in this study was our use of normal nonbrachycephalic dogs in whom the soft palate differs from that of brachycephalic obstructive airway syndrome dogs. The relatively short follow-up could also have impeded the detection of additional events. Finally, the timing of our histologic evaluation made comparison with other studies difficult.
Conclusion
PBRA in cutting mode is a safe technique for resection of the soft palate in nonbrachycephalic dogs. This technique compared favorably with the traditional technique in terms of ease, surgical time, and postoperative inflammation and edema. The lesion created with PBRA was also of limited depth and shallower than that produced by the traditional technique. PBRA can be considered an alternative technique for resection of elongated soft palate in dogs as it ensures better regularity of the palatine wound and more rapid sutureless healing. Future clinical and histologic studies are warranted to determine the nature of the early superficial effect and to validate the effectiveness of PBRA in brachycephalic airway obstructive syndrome dogs.
C-reactive protein results for the plasma-mediated bipolar radiofrequency ablation group (light gray bars) and incisional soft palate resection group (dark gray bars). The upper limit of the reference range is indicated with a horizontal dotted line. Asterisks indicate significant differences between the two groups.
Macroscopic appearance of the soft palate of one representative dog in the PBRA group (above) and INC group (below), immediately after the operation (T0), and on postoperative day 1 (D1), day 3 (D3), day 7 (D7), and day 14 (D14). In each photograph, the tongue is on the bottom and the hard palate is on the top. A discolored area (asterisk) delineated by a thin reddish rim (arrowheads) is visible on the free border of the soft palate, 7 days (D7) after PBRA. INC, incisional soft palate resection; PBRA, plasma-mediated bipolar radiofrequency ablation.
Representative photographs of histological slides. The depth of tissue injury (black line measured in mm) is lower with PBRA (A) than with the traditional incisional technique (B).
Contributor Notes
