- Email: [email protected]
Randomized Study Comparing Equal Height Staples With Graduated Height Staples in Bronchial Closure
Randomized Study Comparing Equal Height Staples With Graduated Height Staples in Bronchial Closure
GENERAL THORACIC
Jiro Okami, MD, PhD, Toshiteru Tokunaga, MD, PhD, Takashi Kanou, MD, PhD, Hidenori Kunou, MD, Daisuke Ishida, MD, PhD, Ayako Fujiwara, MD, Yuri Ito, PhD, and Masahiko Higashiyama, MD, PhD Department of General Thoracic Surgery and Center for Cancer Control and Statistics, Osaka Medical Center for Cancer and Cardiovascular Diseases, Osaka, Japan
Background. There are two forms of staple cartridge available for lung cancer operations, the flat face with equal height staples and the stepped face with graduated height staples. The objective of the study was to evaluate their stapling ability in lobectomy. Methods. A randomized prospective study was conducted to compare two types of stapling reloads for bronchial closure. Stapling ability was evaluated by the staple formation on the resected bronchial stump. The stumps were sliced along the staple line, and the shapes of the staples were recorded by roentgenogram. Then, the staple formation was scored as 0 to 4 points with 4 approximating a perfect B-shaped staple. Results. A total of 61 patients were randomly assigned to the equal height staples (n [ 30) or the graduated height staples (n [ 31). From 61 patients, 183 staple lines, which included 1,144 staples, were evaluated. The case
scores were significantly lower in the equal height staples than in the graduated height staples (2.17 versus 2.88, p [ 0.0003), respectively. The percentage of staples that formed a complete “B” shape was significantly higher in the graduated height staples than in the equal height staples (25.3% versus 10.0%, p [ 0.000). No considerable difference was found between the two groups concerning postoperative complications. No bronchopleural fistula was observed. Conclusions. The graduated height staples had significantly higher scores of the staple formation on the bronchia stump than the equal height staples. From the clinical point of view, both the equal height staples and the graduated height staples showed acceptable performance in the stapling ability.
B
conventional form has a flat face with three rows of equal height staples (EHSs). In contrast, a new form of the reload, which is named Tri-Staple technology (Medtronic. North Haven, CT) and introduced into the market in 2010, is highlighted by a stepped face with three rows of graduated height staples (GHSs). According to the patent description [5], this technology results in graduated compression of the tissue, and the smaller staples at the cut line and the larger staples on the outer third row are designed to minimize tissue trauma. Here, we conducted a single-center randomized study to test the stapling ability of the two different types of cartridge reload in lobectomy for primary lung cancer. To evaluate the stapling cartridges, we scored the staples by their shape on the resected bronchial stump as 0 to 4 points with 4 approximating a perfect B-shaped staple. We then compared staple formation scores and the clinical events between the two forms of cartridge reload.
ronchial closure is a crucial step during lobectomy. The incomplete closure of the bronchial stump may develop into bronchopleural fistula (BPF), which is one of the main surgical complications associated with a longer hospital stay or operative deaths. According to published reports, BPF occurs 0.7% to 5.7% after anatomic pulmonary resection [1]. Therefore, to achieve a safe and reliable bronchial closure is a matter of primary concern for thoracic surgeons. To close the lobar bronchus, surgical staplers are widely used. Among the several different types of staplers available, an articulating-jaw type is the principal choice because of the advantage of accessibility into the thoracic cavity through a small window [2]. Currently, two forms of staple cartridge reload are available for the articulating-jaw stapler, and both are widely used in clinical practice in various fields of operation [3, 4]. A
(Ann Thorac Surg 2017;104:1012–9) Ó 2017 by The Society of Thoracic Surgeons
Accepted for publication Feb 21, 2017. Address correspondence to Dr Okami, Department of General Thoracic Surgery, Osaka Medical Center for Cancer and Cardiovascular Diseases, 1-3-3 Nakamichi Higashinari, Osaka 5378511 Japan; email: okami-ji@mc. pref.osaka.jp.
Ó 2017 by The Society of Thoracic Surgeons Published by Elsevier Inc.
Drs Okami and Higashiyama disclose a financial relationship with Medtronic.
0003-4975/$36.00 http://dx.doi.org/10.1016/j.athoracsur.2017.02.070
Ann Thorac Surg 2017;104:1012–9
Patients and Methods Participants, Design, and Objectives
1013
formation by gross appearance, bleeding or oozing, dehiscence, and air leak under pressure of 10 to 20 cm H2O.
Evaluation of Staple Formation After the operation, bronchial stumps were cut out from the resected lobe and fixed in buffered formalin. The stapled stumps were sliced along the staple line. All pieces were embedded in paraffin and photographed by roentgenogram. An example of an X-ray photograph of a staple line is shown in Figure 2A. The tissue thickness of the stapled stumps were measured with electronic calipers. Staple formation was evaluated and scored with points. If the stapler works properly, the two legs hit the corresponding grooves in the anvil and are bent at four locations (a, b, c, and d in Fig 2B) in each staple to hold the bronchial tissue closed. The ideal shape of a staple is a B-shape shown in Figure 2B. Staple formation was evaluated by the number of proper leg bendings of the four bending locations. Each closed staple was scored as 0, 1, 2, 3, or 4. Examples of each scoring pattern are shown in Figure 2C. Scoring was performed by five thoracic surgeons (J.O., T.T., T.K., D.I., and M.H.) independently under blinded conditions. Staple scores were given for each individual staple (range, 0 to 4) and averaged among the five surgeon scorers. Staple scores were then averaged for the staple-line score, and the case score was the average of all staple scores in each patient.
Staplers
Statistical Analysis
The Endo GIA Ultra Universal staplers (EGIAUSHORT; Medtronic) were used for all patients. For the EHS group, the Endo GIA cartridge reload with a flat face and equal height staples was used for bronchial closure (Fig 1A). The selection of either the blue cartridge for medium tissue (no. 030455; Medtronic) or the green cartridge for thick tissue (no. 030456; Medtronic) was made at the discretion of each surgeon. The staple heights of the blue cartridge and the green cartridge are 3.5 mm and 4.8 mm for all rows, respectively. For the GHS group, the Endo GIA cartridge reload with a stepped face and graduated height staples (Tri-Staple technology) was used for bronchial closure (Fig 1B). The purple cartridge for medium/thick tissue (EGIA45AMT; Medtronic) was used for all patients in the GHS group. The staple heights of this reload were 3.0 mm for the inner row, 3.5 mm for the middle row, and 4.0 mm for the outer row.
It was estimated that this study required 30 patients in each group to detect at least a 0.3-point difference in the scores of the staple formation. The analysis of primary and secondary variables was performed on an intentionto-treat basis. Discrete score data are summarized as median with the 25th and the 75th percentiles in parentheses, because the distribution of some variables were skewed. The statistical significance of categorical variables was evaluated by Fisher’s exact test, whereas either the Student t test or the Wilcoxon’s rank sum test was selected, depending on the type of variables. Interobserver agreement of the scoring system among five scorers was determined by calculating the weighted kappa coefficient [6].
Surgical Procedure Surgical procedures were performed in the same way between the groups. The lobar bronchus was closed and divided with one of the two cartridge reloads depending on the 1:1 randomization assignment. Before lobar bronchial stapling, the circumference of the lobar bronchus was measured.
Evaluation of the Bronchial Stump on the Patient Side Immediately after bronchial stapling, the bronchial stump on the patient side was evaluated in terms of staple
Results Study Population A total of 69 patients were randomly assigned to either group for pulmonary lobectomy. The intention-to-treat population consisted of 61 patients (30 in the EHS group and 31 in the GHS group) (Fig 3). In all patients, primary lung cancer was pathologically diagnosed, and complete resection was performed. Operation time was 187.8 minutes (95% confidence interval [CI]: 177.8 to 197.8 minutes). No significant difference was found between the two groups concerning baseline characteristics (Table 1). Outer circumference of the lobar bronchus was 34.5 mm (95% CI: 32.8 to 36.2 mm). From the surgeons’
GENERAL THORACIC
This is a randomized, prospective, single-center study comparing two forms of cartridge reloads: the flat face with EHSs (the EHS group) and the stepped face with GHSs (the GHS group). The protocol was approved by the institutional review board and registered at University hospital Medical Information Network-Clinical Trials Registry (ID: 000009813) in January 2013. The study was performed between March 2013 and November 2013 at Osaka Medical Center for Cancer and Cardiovascular Diseases (Osaka, Japan). Written and informed consent were obtained from all participants. Inclusion criteria were as follows: indication for upper or lower lobectomy for primary lung cancer, age 20 years or older, and an Eastern Cooperative Oncology Group performance status of 0 or 1. Exclusion criteria were as follows: centrally located tumor, severe comorbidity, and patients with history of ipsilateral thoracic operation. Enrolled patients were excluded intraoperatively when the primary tumor or lymph nodes were invaded or adjacent to the lobar bronchus or main bronchus. The allocation sequence was generated using a computerized block randomization. Primary end point was the performance of staple formation at the lobar bronchial stump. Secondary end points were clinical events during the operation and within the first 30 days. Postoperative complications were classified according to the Clavien-Dindo classification.
OKAMI ET AL STAPLE FORMATION OF BRONCHIAL STUMP
1014
OKAMI ET AL STAPLE FORMATION OF BRONCHIAL STUMP
Ann Thorac Surg 2017;104:1012–9
GENERAL THORACIC
Fig 1. Cross-sectional views of the cartridge reloads and of the resulting tissue interface in the (A and C) equal height staple group and in the (B and D) graduated height staple group.
selection of a suitable cartridge, blue cartridges were used for all patients in the EHS group. Marked calcification of the lobar bronchial cartilage was observed by pathologic examination in 5 patients (8.2%) (3 in the EHS group and 2 in the GHS group). All patients were followed up for 30 days after the operation. One patient was lost to follow-up after 30 days for an unknown reason. Remaining patients were followed up for 3 months or longer after the operation with no delayed bronchial issues recorded. Patients’ medical condition was monitored by chest roentgenogram and blood analysis.
Study End Points: Surgical Outcomes Surgical outcomes are summarized in Table 2. There was no 30-day operative deaths or hospital deaths. Major air leakage from the partial dehiscence of the stump occurred because of apparent staple malformation in 1 patient from the EHS group. A right lower lobectomy had been planned for this patient, but the procedure was switched to a right bilobectomy of the middle and lower lobes. On review of the specimen obtained from this patient it was considered that the staple failure was caused by firing the stapler over a calcified bronchial cartilage of the bifurcation between the B6 and the basal bronchial branch. In 2 patients from the EHS group, a few
bubbles were observed from the staple line during an air inflation test under pressure of 10 to 20 cm H2O. Operation was completed as planned after the staple line was oversewn with sutures to repair the air leaks. Of these 3 patients with major or minor air leaks all were men with a smoking history. Bleeding or oozing of the bronchial stump on the patient side was observed in 5 patients (2 in the EHS group and 3 in the GHS group). They were all recognized as oozing, and hemostasis was easily obtained by contact coagulation with electrocautery. Eleven events of postoperative complication were recorded in 8 patients (arrhythmia in 3 patients, prolonged air leak in 2 patients, pneumonia in 2 patients, and empyema in 1 patient). Concerning grade III, 1 EHS patient experienced empyema after prolonged air leak from lung parenchyma, and 1 patient from the GHS group experienced pneumonia that required bronchoscopic aspiration. No considerable difference was between the two groups concerning the incidence of intraoperative events or postoperative complications. No BPF was observed in this study.
Study End Points: Staple Formation From 61 patients, 183 staple lines were collected. One staple line consisted of 6.25 staples on average (range, 4 to 9 staples). A total of 1,144 staples (550 from the EHS group
Ann Thorac Surg 2017;104:1012–9
OKAMI ET AL STAPLE FORMATION OF BRONCHIAL STUMP
1015
was not significant for the inner row (p ¼ 0.1244) (Figs 5B, 5C, and 5D).
Staple Formation and Clinical Variables
Interobserver Differences of Staple Formation Scoring
Fig 2. Staple formation point scale. (A) An example of X-ray photograph of a staple line. (B) An ideal staple formed in a perfect B-shape. Image provided courtesy of Medtronic. Two legs are bent at the four locations (a, b, c, and d). (C) Examples of scoring patterns.
To validate the scoring system of the staple formation introduced in this study, the interobserver variability was evaluated. Of 1,144 staples, the difference between the highest score and the lowest score among five scorers was 0 in 507 staples (44.3%), 1 in 448 staples (39.2%), 2 in 185 staples (16.2%), and 3 in 4 staples (0.26%). The weighted kappa values of 10 pairs from 5 scorers were 0.9568 to 0.9821. These data indicate excellent interobserver agreement.
Comment and 594 from the GHS group) were scored by five surgeon reviewers based on the above-mentioned scoring protocol. The overall staple score was 2.40 (1.60 to 3.60) with its distribution shown in Figure 4A. Among 1,144 staples, 205 staples (55 from the EHS group and 150 from the GHS group) scored 4 of 4 by all scorers. The percentage of staples scoring 4.0 was significantly higher in the GHS group than in the EHS group (25.3% versus 10.0%, p ¼ 0.000 by Fisher’s test). The overall case score (n ¼ 61) was 2.52 (1.95 to 3.19) with its distribution shown in Figure 4B. The overall case scores were compared between the EHS and GHS groups with scores of 2.17 (1.51 to 2.64) and 2.88 (2.36 to 3.41), respectively. The GHS case scores were statistically superior (p ¼ 0.0003) (Fig 5A). The overall staple-line score was 2.43 (1.92 to 3.20). Staple-line scores were compared between the EHS and GHS groups according to rows with 2.00 (1.68 to 2.60) versus 3.33 (2.34 to 3.60) for the outer row, 2.03 (1.38 to 2.64) versus 3.07 (2.49 to 3.68) for the middle row, and 2.24 (1.60 to 2.64) versus 2.88 (2.36 to 3.41) for the inner row, respectively. Significant differences were found between the two groups for the outer row and the middle row (p < 0.0001), demonstrating GHS superiority, although the difference
Since bronchial stapling was first reported in 1959 and the early 1960s [7–11], it has become a standard procedure during lobectomy worldwide. Surgical staplers allow surgeons to close and divide tissues in a quick and easy manner with less risk of surgical contamination. Today, there is no doubt about their usefulness and convenience, and they are indispensable for operations. However, we can find several reports about complications and unfavorable events associated with staplers [12, 13]. Manufacturers continue to invent new devices and make mechanical modifications to existing devices with the objective of improving patient outcomes. When an articulating-jaw stapler is fired, the layers of tissue between the cartridge and the anvil are subject to high clamping compression forces compared with tissue outside the cartridge. A conventional form of the cartridge reload has a flat face with three rows of EHSs. With this design the compressive force is evenly spread over the tissues within the width of the cartridge (Fig 1C); however, there exists a sharp transition of the compressive forces from the unfastened layers of tissue to the fastened layers. This transition may injure the tissue along the edge of the cartridges by stretching them. To
GENERAL THORACIC
When the association between staple formation and clinical variables were examined by univariate analysis, body mass index (BMI) and tissue thickness of the stapled stump were identified. The case scores were significantly lower (p ¼ 0.018 by Wilcoxon test) in patients with a higher BMI (23.0) than with a lower BMI (<23.0) and in thicker tissue stumps (>2.0 mm) (p ¼ 0.0086 by Wilcoxon test). Other baseline variables such as sex, age, smoking history, clinical stage, tumor location, and outer circumference of the lobar bronchus were not associated with the staple formation. Next, the association between staple formation and intraoperative events was examined. The case scores in 5 patients with oozing on the bronchial stump was 2.53 (1.80 to 3.27). No significant difference (p ¼ 0.8334) was found in case scores between patients with or without oozing. The case scores in EHS patients with an air leak on the bronchial stump were 0.25, 1.19, and 1.51. These values were extremely lower than the case scores in patients without an air leak.
OKAMI ET AL STAPLE FORMATION OF BRONCHIAL STUMP
Ann Thorac Surg 2017;104:1012–9
GENERAL THORACIC
1016
Fig 3. CONSORT flow diagram of the study. (EHS ¼ equal height staple; GHS ¼ graduated height staple.)
address this, a manufacturer invented a new form of cartridge reload characterized with a stepped face with three rows of GHSs. The stepped face distance between an anvil and a staple cartridge increases from a centerline of the cartridge to an outer edge with accompanying graduated staple lengths. The shortest staple is on the inner row to the longest one on the outer row (Fig 1D). The increase in stapled tissue thickness is designed to minimize the traumatic force to the fastened layers of tissue. This technology seems particularly advantageous when stapling compressive and elastic tissue such as lung parenchyma, liver, and the gastrointestinal tract. However, the bronchial wall is a hard and rigid tissue, and it was unknown whether this technology would offer a benefit in bronchial stapling. The bronchial stump on the patient side was observed during the surgical procedure to evaluate stapling ability of cartridge reloads. Major air leak due to incomplete stapling was observed on the stump in 1 patient from the EHS group. Other clinical events were all so trivial that
air leaks and oozing were repaired without difficulty. Incidence of intraoperative events was low, and no significant difference was found between the two groups. BPF or other postoperative complications associated with bronchial stapling was not observed. In this context, both the EHS cartridge and the GHS cartridge showed acceptable performance to close lobar bronchus in clinical practice. Ideally, the incidence of BPF would be compared in a randomized clinical trial to assess if there is any difference in stapling ability between EHS and GHS cartridge reloads; however, because this complication is fortunately rare, it would require a prohibitively large randomized trial to appropriately power the study. For this reason, BPF was not a realistic end point for our clinical trial. The importance of staple formation has long been recognized for secure stapling performance since the initial experience of mechanical bronchial closure. Dr Ravitch, one of the pioneers in mechanical stapling in thoracic operations, stated that the final “B” shape of the closed staple is
Ann Thorac Surg 2017;104:1012–9
OKAMI ET AL STAPLE FORMATION OF BRONCHIAL STUMP
1017
Table 1. Patients’ Background EHS Group (n ¼ 30)
Variable
p Value 0.612a
14 16 68.5 (38–79)
17 14 69 (49–79)
0.608b
23.3 (22.1–24.6) 22.9 (21.3–24.5) 0.645c 0.309a 14 19 16 12 0.508a 26 24 4 7 0.198a 15 21 15 10 0.440a 16 20 14 11 34.6 (31.9–37.3) 34.4 (32.1–36.7) 0.915c 1.99 (1.89–2.10) 1.96 (1.85–2.06) 0.661c
a
Determined by Fisher’s exact test. c test. Determined by t test.
b
Determined by Wilcoxon
BMI ¼ body mass index; CI ¼ confidence interval; height staple; GHS ¼ graduated height staple.
EHS ¼ equal
vital to the success of bronchial stapling [10]. For these reasons, the staple formation was used as a surrogate marker for stapling ability. To make the marker measurable, we scored staple formation by counting the number of proper bendings on four locations on a staple. The case score in the GHS group was significantly higher than the case score in the EHS group. Next, the staple formation was compared according to rows. The staple formation was significantly better in the GHS group than in the ESH group for the outer and middle rows, although a significant difference was not observed
Table 2. Intraoperative Events on the Bronchial Stump of the Patient Side Variable
EHS Group (n ¼ 30)
GHS Group (n ¼ 31)
25 5 1 2 2
28 3 0 0 3
No Yes Major air leak Minor air leak Oozing a
p Valuea
Determined by Fisher’s exact test.
EHS ¼ equal height staple;
GENERAL THORACIC
Sex, n Male Female Age, median (range), years BMI, mean (95% CI) Smoking, n Yes No Clinical stage, n I/II III Side, n Right Left Lobe, n Upper Lower Outer circumference of the lobar bronchus, mean (95% CI), mm Thickness of the stapled stump, mean (95% CI), mm
GHS Group (n ¼ 31)
GHS ¼ graduated height staple.
0.473
Fig 4. (A) The distribution of the staple scores. A total of 1,144 staples were scored by five scorers based on the scoring protocol in the study. (B) The distribution of the case scores of 61 patients. (EHS ¼ equal height staple; GHS ¼ graduated height staple.)
for the inner row. Because the green cartridge was not used in any patient in consequence, this study compared the stapling ability between the blue cartridges and the purple cartridges. The staple height of the middle row in both the blue and purple cartridges are equivalent at 3.5 mm, with the GHS cartridge having an inner row that is 3.0 mm and an outer row that is 4.0 mm, versus the EHS constant staple height of 3.5 mm. Considering that the differences of the staple heights between the groups were 0.5, 0, and 0.5 mm according to the row, the outcomes were not biased by the staple heights. Differences in staple formation could be attributed to the graduated staple heights. As shown in Figure 5C, the staple score of the middle row was significantly higher in the GHS group even though the staple heights were the same in both GHS and EHS (3.5 mm). It was suggested that a stepped face and graduated staple heights contributed to better staple formation in bronchial stapling. The lateral diffusion of compression may result in a higher degree of stapling ability. Two elements may cause staple failure and staple malformation, which are the hardness and thickness of the tissue. If the tissue is hard or thick, the staple legs are
OKAMI ET AL STAPLE FORMATION OF BRONCHIAL STUMP
Ann Thorac Surg 2017;104:1012–9
GENERAL THORACIC
1018
Fig 5. (A) The overall case scores were compared between the two groups. The staple-line scores for the (B) outer row, the (C) middle row, and the (D) inner row. (EHS ¼ equal height staple; GHS ¼ graduated height staple.)
not able to puncture the bronchial wall in the proper direction, and the staples cannot reach the corresponding grooves in the anvil. Consequently, the staples may not form correctly. When univariate analysis was performed to determine whether clinical factors were predictors of staple formation, the thickness of the stapled stumps was a significant factor associated with the
scores. This result supports that the tissue thickness is an important factor associated with stapling performance, and, if the thickness of the clamped lobar bronchus is measured before bronchial stapling, this variable may provide useful information for optimal cartridge selection. BMI was another significant factor associated with the staple scores. Actually, it is difficult to find a
Ann Thorac Surg 2017;104:1012–9
1019
the GHSs showed acceptable performance in the stapling ability. This independent research was supported by a funding contribution from Medtronic (North Haven, CT).
References 1. Sawabata N, Miyaoka E, Asamura H, et al. Japanese lung cancer registry study of 11,663 surgical cases in 2004: demographic and prognosis changes over decade. J Thorac Oncol 2011;6:1229–35. 2. Asamura H, Kondo H, Tsuchiya R. Management of the bronchial stump in pulmonary resections: a review of 533 consecutive recent bronchial closures. Eur J Cardiothorac Surg 2000;17:106–10. 3. Hasegawa S, Nakayama S, Hida K, Kawada K, Sakai Y. Effect of tri-staple technology and slow firing on secure stapling using an endoscopic linear stapler. Dig Surg 2015;32:353–60. 4. Scuderi V, Troisi RI. Tissue management with tri-staple technology in major and minor laparoscopic liver resections. Int Surg 2014;99:606–11. 5. Holsten H, Viola F, Emmons C, Beardsley J, Heinrich R, Cullinan N. Surgical stapling instruments including a cartridge having multiple staples sizes. US Patent 2009; 7588174. 6. Kundel HL, Polansky M. Measurement of observer agreement. Radiology 2003;228:303–8. 7. Amosov NM, Berezovsky KK. Pulmonary resection with mechanical suture. J Thorac Cardiovasc Surg 1961;41:325–35. 8. Androsov PI. New instruments for thoracic surgery. Dis Chest 1963;44:590–7. 9. Goldman A. An evaluation of automatic suture with Ukl-60 and Ukl-40 devices by pulmonary resection. Dis Chest 1964;46:29–36. 10. Ravitch MM, Brown IW, Daviglus GF. Experimental and clinical use of the Soviet bronchus stapling instrument. Surgery 1959;46:97–108. 11. Ravitch MM, Steichen FM, Fishbein RH, Knowles PW, Weil P. Clinical experiences with the Soviet mechanical bronchus stapler (Ukb-25). J Thorac Cardiovasc Surg 1964;47: 446–54. 12. Gossot D, Merlusca G, Tudor A, Boudaya MS, Radu C, Magdeleinat P. Pitfalls related to the use of endostaplers during video-assisted thoracic surgery. Surg Endosc 2009;23: 189–92. 13. Yano M, Yokoi K, Numanami H, et al. Complications of bronchial stapling in thoracic surgery. World J Surg 2014;38: 341–6.
GENERAL THORACIC
reasonable explanation why staple formation was impaired in patients with a higher BMI value. One possible hypothesis is that the bronchial wall might be thicker in the patients with the higher BMI. Further investigation is required for this issue. Marked calcification of the bronchial wall was found in 5 patients in the study. Although the difference did not reach statistical significance, there was a tendency that staple formation was worse in these patients. Overall, it was surprising to find that only a small subset of the staples (only 25% of GHSs and 10% of EHSs) formed into a perfect letter “B” shape when fired on the lobar bronchus. Even though the incidence of adverse clinical events was low in this study, there remains much room for improvement in the stapling ability of surgical staplers pertaining to staple formation. Surgeons and manufacturers should accumulate scientific data on the performance and efficiency of instruments and share these data to promote collaboration to improve the design of the devices and to contribute to reducing unfavorable events. In the interim, based on these results, it is recommended that surgeons should carefully check the staple formation on the bronchial stump and oversew if necessary. There are limitations in this study. First is the assumption that staple formation on the bronchial stump on the patient side is a realistic end point to evaluate stapling ability. Because it is impossible to examine the bronchial stump in situ, the hypothesis is that staple formations are similar on both sides of the cut line. Second, there is no definitive answer to the question of how the staple shape affects clinical outcomes. Because complication rates of BPF and intraoperative air leak are expected to be low, we alternatively defined staple formation scores as a measurable and objective surrogate marker. Third, in addition to GHSs, the new type of cartridge reload also features a fixed and stronger anvil. We cannot exclude the possibility that this technology feature may also play a role to improve the staple formation in the GHS group. In conclusion, the GHSs had significantly higher scores of the staple formation on the bronchia stump than the EHSs. From the clinical point of view, both the EHSs and
OKAMI ET AL STAPLE FORMATION OF BRONCHIAL STUMP
GENERAL THORACIC
Jiro Okami, MD, PhD, Toshiteru Tokunaga, MD, PhD, Takashi Kanou, MD, PhD, Hidenori Kunou, MD, Daisuke Ishida, MD, PhD, Ayako Fujiwara, MD, Yuri Ito, PhD, and Masahiko Higashiyama, MD, PhD Department of General Thoracic Surgery and Center for Cancer Control and Statistics, Osaka Medical Center for Cancer and Cardiovascular Diseases, Osaka, Japan
Background. There are two forms of staple cartridge available for lung cancer operations, the flat face with equal height staples and the stepped face with graduated height staples. The objective of the study was to evaluate their stapling ability in lobectomy. Methods. A randomized prospective study was conducted to compare two types of stapling reloads for bronchial closure. Stapling ability was evaluated by the staple formation on the resected bronchial stump. The stumps were sliced along the staple line, and the shapes of the staples were recorded by roentgenogram. Then, the staple formation was scored as 0 to 4 points with 4 approximating a perfect B-shaped staple. Results. A total of 61 patients were randomly assigned to the equal height staples (n [ 30) or the graduated height staples (n [ 31). From 61 patients, 183 staple lines, which included 1,144 staples, were evaluated. The case
scores were significantly lower in the equal height staples than in the graduated height staples (2.17 versus 2.88, p [ 0.0003), respectively. The percentage of staples that formed a complete “B” shape was significantly higher in the graduated height staples than in the equal height staples (25.3% versus 10.0%, p [ 0.000). No considerable difference was found between the two groups concerning postoperative complications. No bronchopleural fistula was observed. Conclusions. The graduated height staples had significantly higher scores of the staple formation on the bronchia stump than the equal height staples. From the clinical point of view, both the equal height staples and the graduated height staples showed acceptable performance in the stapling ability.
B
conventional form has a flat face with three rows of equal height staples (EHSs). In contrast, a new form of the reload, which is named Tri-Staple technology (Medtronic. North Haven, CT) and introduced into the market in 2010, is highlighted by a stepped face with three rows of graduated height staples (GHSs). According to the patent description [5], this technology results in graduated compression of the tissue, and the smaller staples at the cut line and the larger staples on the outer third row are designed to minimize tissue trauma. Here, we conducted a single-center randomized study to test the stapling ability of the two different types of cartridge reload in lobectomy for primary lung cancer. To evaluate the stapling cartridges, we scored the staples by their shape on the resected bronchial stump as 0 to 4 points with 4 approximating a perfect B-shaped staple. We then compared staple formation scores and the clinical events between the two forms of cartridge reload.
ronchial closure is a crucial step during lobectomy. The incomplete closure of the bronchial stump may develop into bronchopleural fistula (BPF), which is one of the main surgical complications associated with a longer hospital stay or operative deaths. According to published reports, BPF occurs 0.7% to 5.7% after anatomic pulmonary resection [1]. Therefore, to achieve a safe and reliable bronchial closure is a matter of primary concern for thoracic surgeons. To close the lobar bronchus, surgical staplers are widely used. Among the several different types of staplers available, an articulating-jaw type is the principal choice because of the advantage of accessibility into the thoracic cavity through a small window [2]. Currently, two forms of staple cartridge reload are available for the articulating-jaw stapler, and both are widely used in clinical practice in various fields of operation [3, 4]. A
(Ann Thorac Surg 2017;104:1012–9) Ó 2017 by The Society of Thoracic Surgeons
Accepted for publication Feb 21, 2017. Address correspondence to Dr Okami, Department of General Thoracic Surgery, Osaka Medical Center for Cancer and Cardiovascular Diseases, 1-3-3 Nakamichi Higashinari, Osaka 5378511 Japan; email: okami-ji@mc. pref.osaka.jp.
Ó 2017 by The Society of Thoracic Surgeons Published by Elsevier Inc.
Drs Okami and Higashiyama disclose a financial relationship with Medtronic.
0003-4975/$36.00 http://dx.doi.org/10.1016/j.athoracsur.2017.02.070
Ann Thorac Surg 2017;104:1012–9
Patients and Methods Participants, Design, and Objectives
1013
formation by gross appearance, bleeding or oozing, dehiscence, and air leak under pressure of 10 to 20 cm H2O.
Evaluation of Staple Formation After the operation, bronchial stumps were cut out from the resected lobe and fixed in buffered formalin. The stapled stumps were sliced along the staple line. All pieces were embedded in paraffin and photographed by roentgenogram. An example of an X-ray photograph of a staple line is shown in Figure 2A. The tissue thickness of the stapled stumps were measured with electronic calipers. Staple formation was evaluated and scored with points. If the stapler works properly, the two legs hit the corresponding grooves in the anvil and are bent at four locations (a, b, c, and d in Fig 2B) in each staple to hold the bronchial tissue closed. The ideal shape of a staple is a B-shape shown in Figure 2B. Staple formation was evaluated by the number of proper leg bendings of the four bending locations. Each closed staple was scored as 0, 1, 2, 3, or 4. Examples of each scoring pattern are shown in Figure 2C. Scoring was performed by five thoracic surgeons (J.O., T.T., T.K., D.I., and M.H.) independently under blinded conditions. Staple scores were given for each individual staple (range, 0 to 4) and averaged among the five surgeon scorers. Staple scores were then averaged for the staple-line score, and the case score was the average of all staple scores in each patient.
Staplers
Statistical Analysis
The Endo GIA Ultra Universal staplers (EGIAUSHORT; Medtronic) were used for all patients. For the EHS group, the Endo GIA cartridge reload with a flat face and equal height staples was used for bronchial closure (Fig 1A). The selection of either the blue cartridge for medium tissue (no. 030455; Medtronic) or the green cartridge for thick tissue (no. 030456; Medtronic) was made at the discretion of each surgeon. The staple heights of the blue cartridge and the green cartridge are 3.5 mm and 4.8 mm for all rows, respectively. For the GHS group, the Endo GIA cartridge reload with a stepped face and graduated height staples (Tri-Staple technology) was used for bronchial closure (Fig 1B). The purple cartridge for medium/thick tissue (EGIA45AMT; Medtronic) was used for all patients in the GHS group. The staple heights of this reload were 3.0 mm for the inner row, 3.5 mm for the middle row, and 4.0 mm for the outer row.
It was estimated that this study required 30 patients in each group to detect at least a 0.3-point difference in the scores of the staple formation. The analysis of primary and secondary variables was performed on an intentionto-treat basis. Discrete score data are summarized as median with the 25th and the 75th percentiles in parentheses, because the distribution of some variables were skewed. The statistical significance of categorical variables was evaluated by Fisher’s exact test, whereas either the Student t test or the Wilcoxon’s rank sum test was selected, depending on the type of variables. Interobserver agreement of the scoring system among five scorers was determined by calculating the weighted kappa coefficient [6].
Surgical Procedure Surgical procedures were performed in the same way between the groups. The lobar bronchus was closed and divided with one of the two cartridge reloads depending on the 1:1 randomization assignment. Before lobar bronchial stapling, the circumference of the lobar bronchus was measured.
Evaluation of the Bronchial Stump on the Patient Side Immediately after bronchial stapling, the bronchial stump on the patient side was evaluated in terms of staple
Results Study Population A total of 69 patients were randomly assigned to either group for pulmonary lobectomy. The intention-to-treat population consisted of 61 patients (30 in the EHS group and 31 in the GHS group) (Fig 3). In all patients, primary lung cancer was pathologically diagnosed, and complete resection was performed. Operation time was 187.8 minutes (95% confidence interval [CI]: 177.8 to 197.8 minutes). No significant difference was found between the two groups concerning baseline characteristics (Table 1). Outer circumference of the lobar bronchus was 34.5 mm (95% CI: 32.8 to 36.2 mm). From the surgeons’
GENERAL THORACIC
This is a randomized, prospective, single-center study comparing two forms of cartridge reloads: the flat face with EHSs (the EHS group) and the stepped face with GHSs (the GHS group). The protocol was approved by the institutional review board and registered at University hospital Medical Information Network-Clinical Trials Registry (ID: 000009813) in January 2013. The study was performed between March 2013 and November 2013 at Osaka Medical Center for Cancer and Cardiovascular Diseases (Osaka, Japan). Written and informed consent were obtained from all participants. Inclusion criteria were as follows: indication for upper or lower lobectomy for primary lung cancer, age 20 years or older, and an Eastern Cooperative Oncology Group performance status of 0 or 1. Exclusion criteria were as follows: centrally located tumor, severe comorbidity, and patients with history of ipsilateral thoracic operation. Enrolled patients were excluded intraoperatively when the primary tumor or lymph nodes were invaded or adjacent to the lobar bronchus or main bronchus. The allocation sequence was generated using a computerized block randomization. Primary end point was the performance of staple formation at the lobar bronchial stump. Secondary end points were clinical events during the operation and within the first 30 days. Postoperative complications were classified according to the Clavien-Dindo classification.
OKAMI ET AL STAPLE FORMATION OF BRONCHIAL STUMP
1014
OKAMI ET AL STAPLE FORMATION OF BRONCHIAL STUMP
Ann Thorac Surg 2017;104:1012–9
GENERAL THORACIC
Fig 1. Cross-sectional views of the cartridge reloads and of the resulting tissue interface in the (A and C) equal height staple group and in the (B and D) graduated height staple group.
selection of a suitable cartridge, blue cartridges were used for all patients in the EHS group. Marked calcification of the lobar bronchial cartilage was observed by pathologic examination in 5 patients (8.2%) (3 in the EHS group and 2 in the GHS group). All patients were followed up for 30 days after the operation. One patient was lost to follow-up after 30 days for an unknown reason. Remaining patients were followed up for 3 months or longer after the operation with no delayed bronchial issues recorded. Patients’ medical condition was monitored by chest roentgenogram and blood analysis.
Study End Points: Surgical Outcomes Surgical outcomes are summarized in Table 2. There was no 30-day operative deaths or hospital deaths. Major air leakage from the partial dehiscence of the stump occurred because of apparent staple malformation in 1 patient from the EHS group. A right lower lobectomy had been planned for this patient, but the procedure was switched to a right bilobectomy of the middle and lower lobes. On review of the specimen obtained from this patient it was considered that the staple failure was caused by firing the stapler over a calcified bronchial cartilage of the bifurcation between the B6 and the basal bronchial branch. In 2 patients from the EHS group, a few
bubbles were observed from the staple line during an air inflation test under pressure of 10 to 20 cm H2O. Operation was completed as planned after the staple line was oversewn with sutures to repair the air leaks. Of these 3 patients with major or minor air leaks all were men with a smoking history. Bleeding or oozing of the bronchial stump on the patient side was observed in 5 patients (2 in the EHS group and 3 in the GHS group). They were all recognized as oozing, and hemostasis was easily obtained by contact coagulation with electrocautery. Eleven events of postoperative complication were recorded in 8 patients (arrhythmia in 3 patients, prolonged air leak in 2 patients, pneumonia in 2 patients, and empyema in 1 patient). Concerning grade III, 1 EHS patient experienced empyema after prolonged air leak from lung parenchyma, and 1 patient from the GHS group experienced pneumonia that required bronchoscopic aspiration. No considerable difference was between the two groups concerning the incidence of intraoperative events or postoperative complications. No BPF was observed in this study.
Study End Points: Staple Formation From 61 patients, 183 staple lines were collected. One staple line consisted of 6.25 staples on average (range, 4 to 9 staples). A total of 1,144 staples (550 from the EHS group
Ann Thorac Surg 2017;104:1012–9
OKAMI ET AL STAPLE FORMATION OF BRONCHIAL STUMP
1015
was not significant for the inner row (p ¼ 0.1244) (Figs 5B, 5C, and 5D).
Staple Formation and Clinical Variables
Interobserver Differences of Staple Formation Scoring
Fig 2. Staple formation point scale. (A) An example of X-ray photograph of a staple line. (B) An ideal staple formed in a perfect B-shape. Image provided courtesy of Medtronic. Two legs are bent at the four locations (a, b, c, and d). (C) Examples of scoring patterns.
To validate the scoring system of the staple formation introduced in this study, the interobserver variability was evaluated. Of 1,144 staples, the difference between the highest score and the lowest score among five scorers was 0 in 507 staples (44.3%), 1 in 448 staples (39.2%), 2 in 185 staples (16.2%), and 3 in 4 staples (0.26%). The weighted kappa values of 10 pairs from 5 scorers were 0.9568 to 0.9821. These data indicate excellent interobserver agreement.
Comment and 594 from the GHS group) were scored by five surgeon reviewers based on the above-mentioned scoring protocol. The overall staple score was 2.40 (1.60 to 3.60) with its distribution shown in Figure 4A. Among 1,144 staples, 205 staples (55 from the EHS group and 150 from the GHS group) scored 4 of 4 by all scorers. The percentage of staples scoring 4.0 was significantly higher in the GHS group than in the EHS group (25.3% versus 10.0%, p ¼ 0.000 by Fisher’s test). The overall case score (n ¼ 61) was 2.52 (1.95 to 3.19) with its distribution shown in Figure 4B. The overall case scores were compared between the EHS and GHS groups with scores of 2.17 (1.51 to 2.64) and 2.88 (2.36 to 3.41), respectively. The GHS case scores were statistically superior (p ¼ 0.0003) (Fig 5A). The overall staple-line score was 2.43 (1.92 to 3.20). Staple-line scores were compared between the EHS and GHS groups according to rows with 2.00 (1.68 to 2.60) versus 3.33 (2.34 to 3.60) for the outer row, 2.03 (1.38 to 2.64) versus 3.07 (2.49 to 3.68) for the middle row, and 2.24 (1.60 to 2.64) versus 2.88 (2.36 to 3.41) for the inner row, respectively. Significant differences were found between the two groups for the outer row and the middle row (p < 0.0001), demonstrating GHS superiority, although the difference
Since bronchial stapling was first reported in 1959 and the early 1960s [7–11], it has become a standard procedure during lobectomy worldwide. Surgical staplers allow surgeons to close and divide tissues in a quick and easy manner with less risk of surgical contamination. Today, there is no doubt about their usefulness and convenience, and they are indispensable for operations. However, we can find several reports about complications and unfavorable events associated with staplers [12, 13]. Manufacturers continue to invent new devices and make mechanical modifications to existing devices with the objective of improving patient outcomes. When an articulating-jaw stapler is fired, the layers of tissue between the cartridge and the anvil are subject to high clamping compression forces compared with tissue outside the cartridge. A conventional form of the cartridge reload has a flat face with three rows of EHSs. With this design the compressive force is evenly spread over the tissues within the width of the cartridge (Fig 1C); however, there exists a sharp transition of the compressive forces from the unfastened layers of tissue to the fastened layers. This transition may injure the tissue along the edge of the cartridges by stretching them. To
GENERAL THORACIC
When the association between staple formation and clinical variables were examined by univariate analysis, body mass index (BMI) and tissue thickness of the stapled stump were identified. The case scores were significantly lower (p ¼ 0.018 by Wilcoxon test) in patients with a higher BMI (23.0) than with a lower BMI (<23.0) and in thicker tissue stumps (>2.0 mm) (p ¼ 0.0086 by Wilcoxon test). Other baseline variables such as sex, age, smoking history, clinical stage, tumor location, and outer circumference of the lobar bronchus were not associated with the staple formation. Next, the association between staple formation and intraoperative events was examined. The case scores in 5 patients with oozing on the bronchial stump was 2.53 (1.80 to 3.27). No significant difference (p ¼ 0.8334) was found in case scores between patients with or without oozing. The case scores in EHS patients with an air leak on the bronchial stump were 0.25, 1.19, and 1.51. These values were extremely lower than the case scores in patients without an air leak.
OKAMI ET AL STAPLE FORMATION OF BRONCHIAL STUMP
Ann Thorac Surg 2017;104:1012–9
GENERAL THORACIC
1016
Fig 3. CONSORT flow diagram of the study. (EHS ¼ equal height staple; GHS ¼ graduated height staple.)
address this, a manufacturer invented a new form of cartridge reload characterized with a stepped face with three rows of GHSs. The stepped face distance between an anvil and a staple cartridge increases from a centerline of the cartridge to an outer edge with accompanying graduated staple lengths. The shortest staple is on the inner row to the longest one on the outer row (Fig 1D). The increase in stapled tissue thickness is designed to minimize the traumatic force to the fastened layers of tissue. This technology seems particularly advantageous when stapling compressive and elastic tissue such as lung parenchyma, liver, and the gastrointestinal tract. However, the bronchial wall is a hard and rigid tissue, and it was unknown whether this technology would offer a benefit in bronchial stapling. The bronchial stump on the patient side was observed during the surgical procedure to evaluate stapling ability of cartridge reloads. Major air leak due to incomplete stapling was observed on the stump in 1 patient from the EHS group. Other clinical events were all so trivial that
air leaks and oozing were repaired without difficulty. Incidence of intraoperative events was low, and no significant difference was found between the two groups. BPF or other postoperative complications associated with bronchial stapling was not observed. In this context, both the EHS cartridge and the GHS cartridge showed acceptable performance to close lobar bronchus in clinical practice. Ideally, the incidence of BPF would be compared in a randomized clinical trial to assess if there is any difference in stapling ability between EHS and GHS cartridge reloads; however, because this complication is fortunately rare, it would require a prohibitively large randomized trial to appropriately power the study. For this reason, BPF was not a realistic end point for our clinical trial. The importance of staple formation has long been recognized for secure stapling performance since the initial experience of mechanical bronchial closure. Dr Ravitch, one of the pioneers in mechanical stapling in thoracic operations, stated that the final “B” shape of the closed staple is
Ann Thorac Surg 2017;104:1012–9
OKAMI ET AL STAPLE FORMATION OF BRONCHIAL STUMP
1017
Table 1. Patients’ Background EHS Group (n ¼ 30)
Variable
p Value 0.612a
14 16 68.5 (38–79)
17 14 69 (49–79)
0.608b
23.3 (22.1–24.6) 22.9 (21.3–24.5) 0.645c 0.309a 14 19 16 12 0.508a 26 24 4 7 0.198a 15 21 15 10 0.440a 16 20 14 11 34.6 (31.9–37.3) 34.4 (32.1–36.7) 0.915c 1.99 (1.89–2.10) 1.96 (1.85–2.06) 0.661c
a
Determined by Fisher’s exact test. c test. Determined by t test.
b
Determined by Wilcoxon
BMI ¼ body mass index; CI ¼ confidence interval; height staple; GHS ¼ graduated height staple.
EHS ¼ equal
vital to the success of bronchial stapling [10]. For these reasons, the staple formation was used as a surrogate marker for stapling ability. To make the marker measurable, we scored staple formation by counting the number of proper bendings on four locations on a staple. The case score in the GHS group was significantly higher than the case score in the EHS group. Next, the staple formation was compared according to rows. The staple formation was significantly better in the GHS group than in the ESH group for the outer and middle rows, although a significant difference was not observed
Table 2. Intraoperative Events on the Bronchial Stump of the Patient Side Variable
EHS Group (n ¼ 30)
GHS Group (n ¼ 31)
25 5 1 2 2
28 3 0 0 3
No Yes Major air leak Minor air leak Oozing a
p Valuea
Determined by Fisher’s exact test.
EHS ¼ equal height staple;
GENERAL THORACIC
Sex, n Male Female Age, median (range), years BMI, mean (95% CI) Smoking, n Yes No Clinical stage, n I/II III Side, n Right Left Lobe, n Upper Lower Outer circumference of the lobar bronchus, mean (95% CI), mm Thickness of the stapled stump, mean (95% CI), mm
GHS Group (n ¼ 31)
GHS ¼ graduated height staple.
0.473
Fig 4. (A) The distribution of the staple scores. A total of 1,144 staples were scored by five scorers based on the scoring protocol in the study. (B) The distribution of the case scores of 61 patients. (EHS ¼ equal height staple; GHS ¼ graduated height staple.)
for the inner row. Because the green cartridge was not used in any patient in consequence, this study compared the stapling ability between the blue cartridges and the purple cartridges. The staple height of the middle row in both the blue and purple cartridges are equivalent at 3.5 mm, with the GHS cartridge having an inner row that is 3.0 mm and an outer row that is 4.0 mm, versus the EHS constant staple height of 3.5 mm. Considering that the differences of the staple heights between the groups were 0.5, 0, and 0.5 mm according to the row, the outcomes were not biased by the staple heights. Differences in staple formation could be attributed to the graduated staple heights. As shown in Figure 5C, the staple score of the middle row was significantly higher in the GHS group even though the staple heights were the same in both GHS and EHS (3.5 mm). It was suggested that a stepped face and graduated staple heights contributed to better staple formation in bronchial stapling. The lateral diffusion of compression may result in a higher degree of stapling ability. Two elements may cause staple failure and staple malformation, which are the hardness and thickness of the tissue. If the tissue is hard or thick, the staple legs are
OKAMI ET AL STAPLE FORMATION OF BRONCHIAL STUMP
Ann Thorac Surg 2017;104:1012–9
GENERAL THORACIC
1018
Fig 5. (A) The overall case scores were compared between the two groups. The staple-line scores for the (B) outer row, the (C) middle row, and the (D) inner row. (EHS ¼ equal height staple; GHS ¼ graduated height staple.)
not able to puncture the bronchial wall in the proper direction, and the staples cannot reach the corresponding grooves in the anvil. Consequently, the staples may not form correctly. When univariate analysis was performed to determine whether clinical factors were predictors of staple formation, the thickness of the stapled stumps was a significant factor associated with the
scores. This result supports that the tissue thickness is an important factor associated with stapling performance, and, if the thickness of the clamped lobar bronchus is measured before bronchial stapling, this variable may provide useful information for optimal cartridge selection. BMI was another significant factor associated with the staple scores. Actually, it is difficult to find a
Ann Thorac Surg 2017;104:1012–9
1019
the GHSs showed acceptable performance in the stapling ability. This independent research was supported by a funding contribution from Medtronic (North Haven, CT).
References 1. Sawabata N, Miyaoka E, Asamura H, et al. Japanese lung cancer registry study of 11,663 surgical cases in 2004: demographic and prognosis changes over decade. J Thorac Oncol 2011;6:1229–35. 2. Asamura H, Kondo H, Tsuchiya R. Management of the bronchial stump in pulmonary resections: a review of 533 consecutive recent bronchial closures. Eur J Cardiothorac Surg 2000;17:106–10. 3. Hasegawa S, Nakayama S, Hida K, Kawada K, Sakai Y. Effect of tri-staple technology and slow firing on secure stapling using an endoscopic linear stapler. Dig Surg 2015;32:353–60. 4. Scuderi V, Troisi RI. Tissue management with tri-staple technology in major and minor laparoscopic liver resections. Int Surg 2014;99:606–11. 5. Holsten H, Viola F, Emmons C, Beardsley J, Heinrich R, Cullinan N. Surgical stapling instruments including a cartridge having multiple staples sizes. US Patent 2009; 7588174. 6. Kundel HL, Polansky M. Measurement of observer agreement. Radiology 2003;228:303–8. 7. Amosov NM, Berezovsky KK. Pulmonary resection with mechanical suture. J Thorac Cardiovasc Surg 1961;41:325–35. 8. Androsov PI. New instruments for thoracic surgery. Dis Chest 1963;44:590–7. 9. Goldman A. An evaluation of automatic suture with Ukl-60 and Ukl-40 devices by pulmonary resection. Dis Chest 1964;46:29–36. 10. Ravitch MM, Brown IW, Daviglus GF. Experimental and clinical use of the Soviet bronchus stapling instrument. Surgery 1959;46:97–108. 11. Ravitch MM, Steichen FM, Fishbein RH, Knowles PW, Weil P. Clinical experiences with the Soviet mechanical bronchus stapler (Ukb-25). J Thorac Cardiovasc Surg 1964;47: 446–54. 12. Gossot D, Merlusca G, Tudor A, Boudaya MS, Radu C, Magdeleinat P. Pitfalls related to the use of endostaplers during video-assisted thoracic surgery. Surg Endosc 2009;23: 189–92. 13. Yano M, Yokoi K, Numanami H, et al. Complications of bronchial stapling in thoracic surgery. World J Surg 2014;38: 341–6.
GENERAL THORACIC
reasonable explanation why staple formation was impaired in patients with a higher BMI value. One possible hypothesis is that the bronchial wall might be thicker in the patients with the higher BMI. Further investigation is required for this issue. Marked calcification of the bronchial wall was found in 5 patients in the study. Although the difference did not reach statistical significance, there was a tendency that staple formation was worse in these patients. Overall, it was surprising to find that only a small subset of the staples (only 25% of GHSs and 10% of EHSs) formed into a perfect letter “B” shape when fired on the lobar bronchus. Even though the incidence of adverse clinical events was low in this study, there remains much room for improvement in the stapling ability of surgical staplers pertaining to staple formation. Surgeons and manufacturers should accumulate scientific data on the performance and efficiency of instruments and share these data to promote collaboration to improve the design of the devices and to contribute to reducing unfavorable events. In the interim, based on these results, it is recommended that surgeons should carefully check the staple formation on the bronchial stump and oversew if necessary. There are limitations in this study. First is the assumption that staple formation on the bronchial stump on the patient side is a realistic end point to evaluate stapling ability. Because it is impossible to examine the bronchial stump in situ, the hypothesis is that staple formations are similar on both sides of the cut line. Second, there is no definitive answer to the question of how the staple shape affects clinical outcomes. Because complication rates of BPF and intraoperative air leak are expected to be low, we alternatively defined staple formation scores as a measurable and objective surrogate marker. Third, in addition to GHSs, the new type of cartridge reload also features a fixed and stronger anvil. We cannot exclude the possibility that this technology feature may also play a role to improve the staple formation in the GHS group. In conclusion, the GHSs had significantly higher scores of the staple formation on the bronchia stump than the EHSs. From the clinical point of view, both the EHSs and
OKAMI ET AL STAPLE FORMATION OF BRONCHIAL STUMP