Investigation and validation of a novel Endo GIA stapler for canine lung lobectomies

Abstract

The objectives of this study were to validate the 2.0 mm Endo GIA stapler for lung lobectomies and to compare procedural time and air leakage incidence with suture ligation. Sixteen canine cadavers, 18 to 27 kg, were randomly assigned to undergo lung lobectomy of the right middle lung lobe through intercostal thoracotomy, after which suture ligation or the 2.0 mm Endo GIA stapler was used. Procedural time was recorded. Following the lobectomy, the thoracic cavity was filled with fluid. Positive pressure ventilation was used to hold pressure at 20 cm H₂O for 5 minutes. The bronchus was assessed for air leakage as evidenced by visualization of gas bubbles and the ability to maintain positive pressure. Procedural time and air leakage incidence were compared between groups.

By these assessments, the 2.0 mm Endo GIA stapler was successful for lung lobectomies in all cadavers. There was no significant difference (t = −0.856, P = 0.407) in body weight by procedure assignment. Procedural time was significantly shorter (P < 0.0001) using the Endo GIA stapler compared to suture ligation. There was no significant correlation (r = 0.044, P = 0.873) between body weight and procedural time. There were no incidents of air leakage in either group. The 2.0 mm Endo GIA stapler may be used as an alternative device for lung lobectomies in canine cadavers. Smaller staples may provide more complete compression of hilar vessels and bronchi, resulting in reduced hemorrhage, air leakage, and thoracic contamination in cases with infection or neoplasia.

Introduction

Lung lobectomies are performed in human and veterinary medicine for various conditions including neoplasia, bullae and cysts, lung lobe torsions, lung lobe consolidation or abscessation, severe bronchiectasis, trauma, and congenital abnormalities (1). Depending on the underlying pathology, total lobectomies and partial lobectomies have been performed (1). Traditionally, lobectomies were performed with dissection, ligation, and transection of the individual hilar vessels, followed by ligation, transection, and oversewing of the bronchus by hand suturing. Hemorrhage and air leakage are major complications associated with this procedure, and procedural time can be extensive. The use of surgical staples for closures of lobectomies is now common, with the benefits of decreased procedural time, minimal complications, and the ability to remove multiple lesions or lesions that are difficult to resect with traditional techniques (1–3). Additionally, surgical staples can be deployed using a minimally invasive approach (4–8).

The thoracoabdominal (TA) stapler (Medtronic, Minneapolis, Minnesota, USA) is widely used in lung lobectomies in which an open thoracotomy or thoracoscopic-assisted approach is used. The TA stapler deploys 2 or 3 staggered rows of staples to seal tissue, with the side of the device used as a guide for transection of the tissue post-staple deployment. Multiple studies have validated the TA stapler for use in animal lung lobectomies (1,3,8). Another stapling device option is the gastrointestinal anastomosis (GIA) stapler (Medtronic) for partial and total lung lobectomies. The GIA staple cartridge deploys 3 staggered rows of staples on either side of a surgical blade, which transects the tissue as the staples are deployed. The GIA stapler was initially designed for gastrointestinal surgery. Once successfully used in gastrointestinal surgery, the stapler was assessed for use in human and veterinary thoracic surgery. The endoscopic gastrointestinal anastomosis (Endo GIA) stapler (Medtronic) allows for a completely thoracoscopic surgical procedure. Lansdowne et al (4) reported its use in thoracoscopic lung lobectomies, in which the Endo GIA device was successful in 5 of 9 dogs.

When assessing surgical staplers, the size of the staple is important. The height of the staple should be correlated to the thickness of the tissue to be compressed and sealed. In lung lobectomies this includes the hilar vessels and bronchus as well as lung parenchyma. If the staple is too large, hemorrhage or air leakage may result from the lumen of a vessel or bronchus. If the staple is too small, incomplete closure of the staple, and subsequent incomplete compression of the tissue, may occur (1,8). Previously, staple sizes of 2.5 and 3.5 mm have been recommended for lung lobectomies (1,3,8). The TA V3 cartridge (Medtronic) is commonly used. This instrument contains 3 rows of staples 2.5 mm in height, all compressing to 1.0 mm in height. Previously, the smallest Endo GIA cartridge available contained staples 3.0 mm, 2.5 mm, and 2.0 mm in height from the outer row to the inner row. These staples close to heights of 1.25 mm, 1.0 mm, and 0.75 mm, respectively. The novel 2.0 mm Endo GIA staple cartridge (Medtronic) contains 3 rows of staples that are 2.0 mm in height, and each row has a closure height of 0.75 mm. This may be important because the smaller staples could provide for more complete compression of the hilar vessels and bronchus when chosen appropriately for the patient size. This could decrease the incidence of intra- and post-operative hemorrhage and air leakage. Because the stapling device compresses tissue on both sides of the transected tissue, its use could also decrease thoracic contamination in cases in which there is neoplasia or infection. Additionally, use of the stapling device could decrease anesthetic time, surgical time, and hospitalization costs, as well as allow for a completely thoracoscopic surgical procedure given the size of the handle arm and the articulating nature of the arm tip at the cartridge. This staple cartridge has not been validated for use in veterinary thoracic surgery.

The primary objective of this study was to validate the use of the new 2.0 mm Endo GIA stapler cartridge for lung lobectomies in canine cadavers. Secondary objectives included determining the incidence of air leakage and procedural time for lung lobectomy using the novel stapler and comparing that with air leakage and procedural time for suture ligation. Our primary hypothesis was that the 2.0 mm Endo GIA stapler could be used successfully to perform lung lobectomies in canine cadavers. Our secondary hypotheses were that procedural time would be significantly shorter for stapled lung lobectomies using the Endo GIA stapler compared to suture-ligated lung lobectomies, but that there would be no significant difference in the incidence of air leakage between the 2 procedures.

Materials and methods

Sixteen canine cadavers with a weight range of 18 to 27 kg were procured from local shelters. The sample size used for this study was determined using the following a priori information: an alpha = 0.05, a power of 0.8, an expected difference in surgery time of 15 min, and a standard deviation in each group of 10 min (16 lungs, 8 in each group). The dogs were euthanized for reasons unrelated to this study, were acquired fresh, and procedures were performed within 6 h of euthanasia. The cadavers were randomly assigned into 1 of 2 groups. Group 1 was assigned to lung lobectomies performed by suture ligation (SL) and Group 2 was assigned to stapled lung lobectomies (ST). Cadavers with gross pulmonary parenchymal pathology were excluded from the study population.

All dogs were placed in left lateral recumbency and intubated via a tracheostomy, with the cuff of the endotracheal tube inflated so that no leak was detected with positive pressure ventilation at 25 cm H₂O before the beginning of surgery. All dogs underwent a right lateral thoracotomy at the 5th intercostal space. A Finochietto rib retractor was used to aid in exposure of the right hemithorax. The 8 dogs in Group 1 (SL) underwent a right middle lung lobectomy using dissection, ligation, and transection of the individual hilar vessels and bronchus. The vessels were triple ligated separately, with 2 encircling ligatures of 2-0 monofilament absorbable suture (Maxon; Medtronic) proximally and distally, and 1 transfixation ligature in between. The vessels were transected between the 2 distal ligatures. The bronchus was ligated with 3 encircling ligatures using the same suture and was transected between the 2 distal ligatures. The bronchus stump was oversewn with 4-0 monofilament absorbable suture (Biosyn; Medtronic). The dogs in Group 2 (ST) underwent a right middle lung lobectomy using the novel 2.0 mm Endo GIA stapler. The hilar vessels and bronchus were identified, and the stapler was deployed perpendicular to the hilus. Once the assigned procedure was completed, the thoracic cavity was filled with fluid. Positive pressure ventilation was used to hold pressure at 20 cm H₂O for 5 min. The submerged bronchus was assessed for air leakage as evidenced by gas bubbles within the fluid. Inability to maintain positive pressure at 20 cm H₂O was also considered to be air leakage from the lung. All procedures were performed by a single surgeon: a surgery resident in an American College of Veterinary Surgeons (ACVS)-approved small animal surgery training program.

Data collected included procedural time, presence or absence of gas bubbles, and ability to maintain positive pressure ventilation at 20 cm H₂O. Procedural time started after the thoracotomy was completed and the Finochietto rib retractor was placed, and as dissection around the hilus began for both groups. Time of surgery ended when the bronchus was oversewn in Group 1 (SL), and when the stapler was fired and released in Group 2 (ST), but before the thoracic cavity was filled with fluid. Presence or absence of air leakage was determined by the presence or absence of gas bubbles visualized when positive pressure ventilation was held at 20 cm H₂O for 5 min as well as maintenance or drop in positive pressure from 20 cm H₂O.

The distribution of the data was evaluated using the Shapiro-Wilk test, skewness, kurtosis, and q-q plots. Body weight was normally distributed and was reported by the mean, standard deviation, and minimum–maximum values. Time for the procedure was not normally distributed. Attempts to log transform the data were not successful, thus the data are reported by the median, 25 to 75 percentiles, and minimum–maximum values. An independent samples Student’s t-test was used to determine if there was a significant difference in body weight by closure technique assignment. Levene’s test was used to test for equality of variance. A Mann-Whitney U-test was used to determine if there was a significant difference in procedural time between closure techniques. Pearson’s correlation test was used to determine if there was an association between procedural time and body weight. All analyses were conducted using commercially available software (SPSS Statistics for Windows, version 24.0; IBM Statistics, Armonk, New York, USA). A P < 0.05 was used to determine statistical significance.

Results

The novel 2.0 mm Endo GIA stapler was used successfully in all cadavers assigned to Group 2 (ST). Body weight appeared to be evenly distributed, as there was no significant difference (t = −0.856, P = 0.407) in body weight between groups (SL, mean: 21.3, SD: 2.1, min–max: 19.3–25.4; ST, mean: 22.4, SD: 2.6, min–max: 18.6–25.4). Procedural times and air leakage are reported in Table 1. Procedural time was significantly shorter (P < 0.0001) in lung lobectomies using the novel stapler (mean: 2.87, median: 2.70, SD: 0.96, 25 to 75%: 2.14 to 3.53, min–max: 1.55–4.57) compared to suture ligation (mean: 23.46, median: 21.75, SD: 8.22, 25 to 75%: 18.67 to 22.42, min–max: 17.65–43.3). There was no significant correlation (r = 0.044, P = 0.873) between body weight and procedural time. There were no incidents of air leakage in either group.

Table 1.

Demographics of study subjects and outcomes of surgical procedures.

Case number	Body weight (kg)	Group	Air leakage pressure (at 20 cm H2O)	Procedure time (min)
1	25.45	1 (SL)	None	43.3
2	20.27	1 (SL)	None	22.15
3	20.18	1 (SL)	None	22.2
4	23.82	1 (SL)	None	22.5
5	19.27	1 (SL)	None	21.35
6	25.36	2 (ST)	None	2.53
7	23.55	2 (ST)	None	4.57
8	18.64	2 (ST)	None	2.05
9	24.18	2 (ST)	None	3.47
10	21.91	2 (ST)	None	3.55
11	21.18	1 (SL)	None	17.65
12	20.91	1 (SL)	None	18.03
13	19.73	1 (SL)	None	20.57
14	21.82	2 (ST)	None	2.88
15	18.73	2 (ST)	None	1.55
16	24.82	2 (ST)	None	2.42

SL — Suture-ligated; ST — Stapled.

Discussion

Complete lung lobectomies were performed successfully in all cadavers between 18 to 27 kg using the novel 2.0 mm Endo GIA stapler, thus supporting our primary hypothesis. Procedural time was significantly shorter for stapled compared to suture-ligated lung lobectomies. There were no incidents of air leakage in either group; therefore, our secondary hypotheses were likewise supported. Size was not a limitation within this study population of dogs, as all procedures were performed successfully.

To the authors’ knowledge, there is only 1 prospective study comparing pain and morbidity in dogs undergoing open thoracotomies versus thoracoscopic procedures. In that study, dogs undergoing thoracoscopic pericardectomies had less pain, fewer wound complications, and returned to function more quickly (9). Considering that many surgeons and owners might prefer minimally invasive approaches for these reasons, it is important to ensure that the stapling device options used for thoracoscopic lung lobectomies are both effective and safe. The cohort of dogs used herein (18 to 27 kg) was chosen to validate the use of this novel staple cartridge in a group of medium- to large-sized dogs. Although staple sizes of 2.5 and 3.5 mm have been recommended (1,3,8), to the authors’ knowledge, no studies have evaluated the ideal staple size for dogs which vary greatly in size. Smaller staples have a theoretical advantage in small-sized canine and feline populations, allowing a more appropriate staple-to-tissue size match. This may allow more complete compression of the vessels and bronchus at the hilus and further limit air leakage and hemorrhage. In addition, the articulating arm of this device is also advantageous in terms of maneuverability in the small working space in these patients. This benefit is not unique to total thoracoscopic surgery and is applicable to thoracoscopic-assisted and open procedures as well. Future studies are needed to investigate the use of this staple cartridge in small-sized dogs and cats.

One cadaver in Group 1 (Case 1) had 3 pulmonary veins to dissect and ligate individually, prolonging procedural time. When this outlier was excluded, the procedural time of lung lobectomies using the stapler was still significantly shorter than that of lung lobectomies using suture ligation (P < 0.0001). This case demonstrates the variable anatomy of pulmonary vasculature, which can prolong procedural time with hand suturing but is of no consequence using a stapling device because all vessels are ligated and transected simultaneously. Increased experience did not seem to increase efficiency with the use of this stapler, as the procedural times were similar regardless of the number of lobectomies previously completed. In addition, the stapler was subjectively easy to use.

This study had several limitations. All lung lobectomies were performed by a single surgeon: a small animal surgery resident in training. Relative inexperience may have led to artificial prolongation of the procedural time, particularly in the sutured lung lobectomies. The relative inexperience of the surgeon suggests that the findings of this study may be particularly relevant to less experienced surgeons. However, increased experience did not seem to increase efficiency with the use of the stapler, as procedural times, except for the outlier previously described, were similar among the remainder of the cadavers in the SL group. It is possible that there was an insufficient number of procedures to identify a difference for either the sutured or the stapled procedures.

All lobectomies were performed on a single lung lobe (right middle lobe) in an attempt to standardize the approach and procedure. It is unlikely that the use of the stapler on other lobes would have changed the incidence of air leakage, although the procedural times may have been longer on lobes requiring dissection of the pulmonary ligament or in lungs with pathology such as neoplasia or infection, and with the presence of adhesions.

Cadaveric work has inherent limitations. Cadaveric tissue, however fresh, may not exactly mimic tissue of a live patient, especially in terms of its ability to form a seal at the lobectomy site. There is also absence of blood flow, which prevents evaluation of hemorrhage at the suture or staple line. In a live patient, the pulsatile nature of blood vessels presents an additional challenge to the lobectomy site. Pulsation of blood exerts pressure on the suture or staple line which may lead to hemorrhage that would not be identified in a cadaver. The short-term nature and use of cadavers in the study also prevented assessment of slow air leakage and long-term thoracic contamination at the staple line. Therefore, it was not possible to identify patients in which incomplete closure of the tissue would have led to bacterial contamination or contamination with neoplastic cells in cases with infection or neoplasia. Future investigations should explore the use of the 2.0 mm Endo GIA stapler for open and thoracoscopic lung lobectomies in live veterinary patients to ensure its ability to achieve adequate hemostasis and to prevent slow air leakage and thoracic contamination. The 2.0 mm Endo GIA stapler has already been used in clinical patients by the authors and no complications were noted in the post-operative or at follow-up appointments.

In conclusion, the 2.0 mm Endo GIA stapler can be used successfully to perform lung lobectomies in canine cadavers. Procedural time was significantly shorter for stapled lung lobectomies using the 2.0 mm Endo GIA staple cartridge compared to suture-ligated lung lobectomies. Lastly, there were no incidents of air leakage in any of the procedures. The 2.0 mm Endo GIA stapler provides surgeons with an alternative, smaller cartridge size to use for lung lobectomies.