The physiological ventilatory pressure considered in this study was 20 cmH2O (15 mmHg). According veterinary guidelines, pressure levels on lung tissue from 15 to 25 cmH2O can be considered physiological13. However, pressure values of 15 to 20 cmH2O are considered more accurate for small animals under normal respiratory mechanics conditions14. In one study, aerostasis was considered safe when air leakage occurr at ventilatory pressures greater than 20 cmH2O after a lung biopsy in anesthetised dogs11. In another experimental study evaluating different types of surgical staplers, air leakage at a ventilatory pressure equal to or less than 20 cmH2O was considered not safe and the sealing device inadequate12. Sealing methods probably should withstand a ventilatory pressure of the airways of up to 20 cmH2O to be considered safe. All the methods used for sealing that were investigated in this study were effective against established physiological ventilatory pressure.
The canine lung model used wasmacroscopically normal butit was obtained from dogs with leishmaniasis. Most of the dogs were clinically unwell which resulted in different body conditions score between the groups. The body condition score values15 of G3 were lower (1–3), compared to G1 (3–7). However, there was not a statistical difference in the aerostatic property of these groups. Despite difference between median lung lobe height of G4 (Md = 12.12 cm) and G5 (Md = 19.47 cm), G5 showed superiority sealing property under supraphysiological pressures compared to the median obtained by G4, which proves that even with a larger surface area to be sealed, the synthetic glue was still able to withstand against high pressures ventilation, ensuring greater aerostatic safety than the staples device.
One of the manual sutures used and analyzed in this study was the cobbler suture16 (Fig. 3A) which ensured good sealing in almost all the lung lobes (Fig. 3B). However one developed air leakage, at the pressure of 14 mmHg (= 19.03 cmH2O). The punctual air leak detected in the place where the suture was contained, possibly corroborates the observation described by other authors when evaluating the double-layer mattress suture in sheep lungs after partial resection of one of the lobes10. The authors observed minimal air leakage along the parenchyma when the lungs were inflated to 40 mmHg and explained that this was due to the portion of the parenchyma along the needle incisions. Thus, the small leak detected in 1/6 model after cobbler suture probably occurred when the needle passed through the tissue. However, there were no statistical differences between the models analyzed in the present study. The successful aerostasis observed in this group can be attributed to the increase in tissue contact surface caused by this suture pattern, which ensured sufficient sealing against physiological ventilatory pressures.
The overlapping continuous suture (Fig. 4A) used in G2 has been described for lung sealing after partial lobectomy in dogs and cats17,18. This is a manual suture pattern defined as a pneumostatic suture, once it results in additional compression of the parenchyma 17. The results found in this study confirm the efficacy of this suture patternin maintain aerostasis against physiological ventilatory pressures (Fig. 4B).
The third manual suture investigated (G3) was the Ford interlocking suture pattern (Fig. 5A), characterized by a modified continuous suture and frequently used in soft tissue surgeries in order to withstand some degree of tissue tension19. Such sealing suture also demonstrated a good aerostasis, although two of the models showed air leakage at physiological ventilatory pressures. It is worth mentioning that so far, there are no references in the veterinary literature regarding the use of the Ford interlocking suture pattern in loborraphy after partial lung lobectomy in dogs. This suture pattern is indicated for the correction of lacerations of the lung parenchyma in humans, since it promotes good adhesion of its edges, and therefore hemostasis20. This suture brings the margins closer providing an effective seal. Despite the lack of studies with the Ford interlocking suture technique in dogs in vivo, this sealing tecnique could be a safe option after partial pulmonary lobectomy in dogs. Two out of six cadaver models in the Ford interlocking suture group failed to achieve aerostasis when the ventilatory pressure reached a value of 19.03 cmH2O (Fig. 5B). In both cases, along the suture line, there was no leakage, except for a single laceration point lateral to the suture line. When the ventilatory pressure wasincreased, the tissue tension in the lobe was probably higher in certain points, which may have caused a punctual laceration, and consequent leakage of air19,21,22.
In the group of lung lobes in which the mechanical suture was performed with metal staples using the TA linear surgical stapler, the results showed a good response in terms of aerostasis, with no difference compared to the other groups. However two out of six cases failed due to air leakage laterally to the staple line, at physiological ventilatory pressures of 13.59 cmH2O and 19.03 cmH2O. The effective aerostasis with mechanical stapling sealing devices and observed in the present experimental study was also reported by other authors2,23. The safe use of staples has been previously demonstrated in nearly 40 dogs and cats undergoing pulmonary lobectomy for the treatment of different conditions2. In this study only two patients developed pneumothorax in the postoperative period2. In an ex vivo study with lung biopsy in dogs, airway leakage occurred at pressure of 28 cmH2O. However this was achieved using an EndoGIA 45 mm endoscope stapler11. In another study, also using a stapler during thoracoscopy, eight dogs were submitted to partial and total pulmonary lobectomies, and although the pulmonary inflation pressure was not mentioned, all animals were discharged without intraoperative complications after loborraphy with surgical staples24. In a clinical study two main conventional methods of pulmonary sealing were compared after partial lobectomy in eight dogs. It was found that the mechanical suture with staplers was superior over manual suture with non-absorbable material23. The authors concluded that staples could be superior as it achieves an equal distribution along the parenchyma, while manual suture, are more likley to depend on the surgeons tecnique which may lead to variability in tissues sealing distribution23. Furthermore, the efficacy of surgical staples over manual sutures has been proven in humans undergoing pulmonary lobectomy25. The authors also believe that manual suturing depends on the surgeon's experience and skills, and manual suturing could cause more air leakage caused by the passage of the needle throughout the pulmonary parenchyma25. However, in the present study, there was no statistical difference regarding the efficacy of aerostasis between manual and mechanical sutures. However, in the present study, aerostasis promoted by mechanical suturing was achieved by ensuring the insertion of staples for all the lenght of the sectioned parenchyma 6. To avoid air leakage due to failure of inserted staples, two blue cartridges with a total width of 45 mm each were used, containing a double row of staples (Fig. 6A). The width x height dimension of each of these staples when opened was 4.0 x 3.5 mm. Both cartridges were inserted in opposite directions in the lobe, in order to ensure that the lateral ends of the parenchyma to be stapled were all completely covered by staples, thus avoiding air leakage from uncovered areas. The presence of a wider cartridge, such as the 60 mm one, could haved replaced the use of the two cartridges used in the study.
An interesting observation is that the leg length of the staple in mm did not match the thickness of the tissue to which it was inserted. According to the type of surgical stapler used in this study, the blue cartridge of the TA linear stapler has a leg length of 3.5 mm when open, with a staple height of 1.5 mm when closed26. The veterinary literature recommends that when surgical staples are used in the lung parenchyma, the selection of the height of the staples to be used depends upon the compressed thickness of the tissue6. The discrepancy of the height in vivo should be taken in consideration before deciding the type of staple to apply. However in our study lung lobes with different values of height or parenchyma thickness were used. In the stapling device group, the thinnest parenchyma was 10.89 mm and the largest was 14.33 mm, which means that these values were higher than the standardized heights of the staples used in the study. Despite an incomplete cover of the staple over all the parenchyma to be sealed, air leakage did not occur, since the six lobes of G4 did not show differences at physiological pressures. Probably, the reduced length of the staple when closed, should consider only the diameter of the bronchioles and not the all thickness of the tissue to ensure aerostasis6. However, the use of staples in pulmonary loborrhaphy in dogs, although considered safe, does not rule out completely the risk of air leakage during intraoperative positive pulmonary ventilation as observed in the two out of six models in the staples group. Air leakage has been documented at the staples site under physiological ventilatory pressures, in other studies with dog’s lung12. In a study evaluating three different models of surgical staplers used in the pulmonary loborrhaphy of canine cadavers, the rupture pressure at the stapling line seldom occurred at values equal to or less than 20 cmH2O12. Air leakage along the stapling line was reported by other authors when using an EndoGia 45 mm endoscopic stapler. When performing lung lobe biopsy in vivo in dogs and comparing Endo Gia 45 mm with vascular sealing devices and sutures, air leak was found at the stapler site and possibly caused by staple trauma to the parenchyma. The authour concluded that the stapler device was considered safe because the median pressure values of air leakage was not less than 20 cmH2O11. At high positive pressure he inserted staples can lacerate the lung tissue, resulting in wider orifices and consequent air leakage9 even at pressures around 20 to 25 cmH2O. This is the likely explanation for what may have caused air leackage in the two models in the staples group (Fig. 6B). Although the superiority of mechanical suturing has been reported, an additional manual suture over the staple line used for partial pulmonary loborrhaphy may be necessary to control parenchymal leakage and bleeding23. This suture reinforcement in the stapling line is also recommended by other authors, and can be performed with additional stapling or with manual suturing over the area of air leakage6,26. Complementing the staple line with other surgical suture or devices could guarantee better resistance at supraphysiological levels of ventilatory pressure8,27. However this was not evaluated in our study.
The effective sealing property provided by the Glubran-2® surgical sealant (G5) did not result in aerostasis failure in any of the six lungs when challenged by physiological ventilatory pressure. Air leakage occurred only at very high ventilatory pressure, with a minimum of 20 mmHg (or 27.19 cmH2O) and a maximum of 80 mmHg (or 108.76 cmH2O) (Figs. 7AB). Similar results were achieved in an experimental study using rabbit lungs and sealant that contains NBCA, one of the monomers present in Glubran-2®. In such study with rabbits the authors demonstrated that of the 12 lung lobes resection, air leakage developed only at supraphysiological ventilatory pressures, ranging from 18.38 to 40.45 mmHg (25 to 55 cmH2O)28. Similarly to our study sealing failure occurred only at pressure above 27 cmH2O, demonstrating greater safety after pulmonary loborrhaphy in dogs. In the present study, two of the six lobes in the glue group had aerostasis failure, at 60 and 80 mmHg, 4–5,.4 times higher than the physiological pressure, and above the maximum pressure recorded in the study with rabbit lungs28. Other similar result was reported in an experimental study that used the same type of sealant in 20 sheeps lungs against supraphysiological pressures of 40, 60 and 80 mmHg10. The authors noted that most of the lobes did not present air leakage in the region sealed by the glue, except for a smaller percentage of 20% of the lobes that resulted in minimal degree of air leackage.
The Glubran-2® sealant is one of the commercial names of NBCA, an important synthetic biodegradable surgical glue, which is modified by the addition of MS monomer. It has been widely used in medicine, such as in urological, gynecological, digestive, neurological, cardiovascular and thoracic surgeries because it demonstrates adhesive, hemostatic and aerostatic properties when applied to diferent tissues29− 31. Safety and efficacy of this synthetic glue has been reported in cats, swine, rabbits, sheeps and humans10,28,32− 35. However, so far, no reports have been described or published using the cyanoacrylate co-monomer Glubran-2® in surgeries involving lobectomy in dog’s lungs. In a study with six healthy cats after performing partial pulmonary lobectomy, NBCA has effective hemostatic and aerostatic property32. The results were satisfactory, although the ventilatory pressure used to assess aerostasis was not reported, nor the extent of resected parenchyma. However, due to the absence of clinical significant post-surgical complications and the histopathological evaluation of the patches containing the glue three weeks after surgery, it was concluded that this type of cyanoacrylate would be a safe alternative method for pulmonary loborrhaphy in cats32. From radiographic examinations, mild asymptomatic pneumothorax was observed only in 4 patients soon after surgery and in 1 patient two days post surgery 32. The sealing property of the glue is due to the formation of a thin elastic film of high strength that molds itself over the tissue, and makes the tissue region firm within 60 to 90 seconds. This is the result of the glue exothermic polymerization after interacting with tissue surface, forming strong bonds with tissue proteins and ensuring strong sealing to the lung tissue at the place where it is applied29,34,36. In the experimental study with rabbit lungs local and systemic effects were investigated after the use of the synthetic cyanoacrylate sealant. It was macroscopically observed that the portion of the lung parenchyma that was glued had an altered texture and hardened consistency for almost a month after being applied. This highlights the efficacy of the sealing property of this product for pulmonary aerostasis36. In the current study, although no difference was observed between the groups regarding rupture under physiological pressure, the sealing promoted by the synthetic glue containing NBCA was able to ensure an efficient aerostasis also at supraphysiological ventilation pressures. The median pressure of rupture in G5 (Glubran-2®) was superior to observed in G3 (Ford interlocking suture) and G4 (staples).
In the present study, the canine lung lobes of the 3 groups that received the different manual sutures made with the same type of surgical material, were not able to achieve aerostasis at pressure levels of 40 mmHg,. Similarly, as demonstrated in the study carried out with sheeps lungs after partial lobectomy, all lobes that received manual sutures failed aerostasis after applyng this pressure10. However, among the manual sutures analyzed in the current study, only the group that received the Ford interlocking suture showed a significantly lower difference compared to the group that received the synthetic glue. The air leak observed in the manual suture line proved to come from the place where the needle passed in the parenchyma10. Similar to our study, the authors concluded that the aerostasis promoted by the sealing performed by the manual suture, was inferior to the sealing performed by the synthetic glue, in all 20 sheeps lungs evaluated at high ventilatory pressure10.
In our study synthetic cyanoacrylate adhesive was significantly superior to the manual Ford interlocking suture, and to mechanical suturing by staples. The excellent seal maintained by the synthetic glue can be explained by the low viscosity of the sealant when in liquid phase filling the small gaps and creating better sealing between the cutting surfaces of the lung than other sealing methods29. The glue also does not create physical injury to the lung parenchyma, as the tissue trauma caused by the needle and staples10,11,25.
The main limitation of this study is the experimental model. The lungs were evaluated outside the thoracic cavity, a condition that prevents adequate lung distension and retraction during the respiratory cycle37. Under laboratory conditions accurate assessment of the sealing methods against physiological variations of ventilatory pressures is not completely possible. On post mortem examination the lungs appeared macroscopically normal and the experiment carried out within a few hours after death. However this is not completely comparable to a lung in a physiological condition. Therefore, the safety and eficacy data of aerostasis in all the methods evaluatedin this experimental model, should not be completely and safely extrapolated for dogs in vivo. Another limitation regards the collection of lungs from patients, diagnosed with leishmaniasis. It is possible that microscopic lung lesions could be present at the time of the study, without being macroscopically visible. Leishmania sp is known to cause histological changes including mild to severe peribronchial inflammatory infiltration, fibroblast proliferation in lung tissues, edema and congestion in the alveolar wall, but histological examination of the lungs in this study were not performed38. Microscopic changes, whether caused by Leishmania sp or some subclinical respiratory disease, could interfere with lung compliance and adequate sealing, resulting in rupture at lower pressures, and could justify some results of this study7,11,38.