Lasers for Pulmonary Resections: A Qualitative Systematic Review
Objective
Our goal is to evaluate the effectiveness and safety of lasers compared with mechanical staplers in pulmonary resections using a qualitative and systematic manner.
Summary of background data
The laser technique is effective in pulmonary resections among patients with pulmonary disease. However, systematic review of this technique needs to be made to ascertain its key role in the thoracosurgery domain.
Methods
For randomized controlled trials (RCTs) published in Medline, Web of Science, Cochrane Controlled Trials Register, and clinical trial databases from June 1979 to October 2022, we identified and synthesized 3 studies to compare the efficacy and safety of lasers and mechanical staplers in pulmonary resections based on selection criteria. Two reviewers independently assessed trial bias and extracted data to make a useful qualitative systematic review.
Results
The 3 RCTs were obtained using a surgery approach of video-assisted thoracoscopic surgery. The operating time of the laser group was longer than that of the staple group in the Marulli study [weighted mean difference (WMD) = 12.00 minutes, 95% confidence interval (CI) -11.66 to 35.66 for McKenna; WMD = 39.00 minutes, 95% CI 21.82 to 56.18 for Marulli; and WMD = 2.00 minutes, 95% CI -15.10 to 19.10 for Scanagatta], whereas the hospital stay of the laser group was comparative with that of the staple group (WMD = -2.00 days, 95% CI -7.36 to 3.36 for McKenna; WMD = -3.00 days, 95% CI -6.29 to 0.29 for Marulli; and WMD = 0.00 days 95% CI -1.69 to 1.69 for Scanagatta). The risk ratio [RR (95% CI)], expressed as the persistent air leaks of the laser group versus the staple group, was RR = 0.68 (95% CI 0.38 to 1.22) for McKenna, RR = 0.67 (95% CI 0.12 to 3.61) for Marulli, and RR = 1.07 (95% CI 0.53 to 2.16) for Scanagatta, respectively, and expressed as pneumothorax, RR = 7.09 (95% CI 0.90 to 55.95) for McKenna, RR = 0.33 (95% CI 0.01 to 7.76) for Marulli, and RR = 6.28 (95% CI 0.34 to 117.39) for Scanagatta, respectively. At the 6-month follow-up, the mean postoperative forced expiratory volume in 1 second of the staple group was significantly improved compared with that of the laser group. The clinical symptoms and dyspnea index were improved by more than 1 grade 8 of 33 (24%) patients in the laser group and 26 of 39 (66%) patients in the staple group (p = 0.003).
Conclusions
The lasers are effective and comparable with the mechanical stapler technique in pulmonary resections.
The field of resection for patients with malignant lung diseases and some nonmalignant conditions (emphysema, pulmonary sequestrations, abscesses, bronchiectasis, and chronic pulmonary infections with localized parenchymal disruption) has attracted more and more attention. Electrocautery is widely used for pulmonary parenchymal resections, but persistent air leaks and pneumothorax often occur, and additional tools, including mechanical staplers and lasers, have been introduced to reduce these complications. In recent decades, more and more surgeons have generally accepted mechanical staplers to cut and suture pulmonary tissues in anatomic lung resections.1 Especially in laparoscopic pulmonary resections, surgical staplers are fired in a staggered formation simultaneously to cut and ligate vessels and bronchus using a 2-dimensional planar mode.
Since the 1960s, lasers, defined as light amplification by stimulated emission of radiation, have been considered a preferable choice for a variety of medical operations, including coagulation, evaporation, and cutting of surgical tissues.2 By the 1980s, various types of lasers had been introduced to a range of thoracic surgical operations and most commonly presented in pulmonary metastasectomies.3,4 In the 1990s, laser bullectomy was employed by Wakabayashi and coworkers for patients with diffuse emphysema.5 In the following years, the application of lasers (in particular 1318 nm neodymium-YAG lasers or 1908 nm thulium lasers) achieved illustrious hemostasis and air-sealing effects on pulmonary resections.6 On the other hand, lung parenchymal tissue contains 80% liquid components and few parenchymal tissue components, so these organs are suitable tissues for lasers.7
Little information is available about the efficacy and safety of lasers used in the surgical treatment of pulmonary disease. The current published articles for lasers used in pulmonary resections are mostly focused on case series and retrospective trials. It is not clear whether these patients who undergo pulmonary resections will benefit most from lasers, and the optimal patient selection criteria are unclear. Although several prospective studies compared these different laser techniques with mechanical staplers in pulmonary resections, the wide heterogeneity of patient selection and assessment of outcomes in these published literatures results in a lack of sufficient evidence for their benefits and harms, which is the reason why we carried out this review. In addition, the efficacy and safety of these 2 kinds of pneumonectomy are still controversial. Based on these trials, we compared the current efficacy and safety of these 2 surgical applications through reviewing randomized controlled trials (RCTs) in a narrative and systematic manner.
Material and Methods
Search strategy
We attempted to identify publications of all relevant studies in Medline, Web of Science, Cochrane Controlled Trials Register, and clinical trial databases between June 1979 and October 2022 using the following search strategies: ((((lung[MeSH Terms]) OR (lung[Title/Abstract])) OR (lungs[Title/Abstract])) OR (pulmonary[Title/Abstract])) AND ((((lasers[MeSH Terms])) OR (laser[Title/Abstract])) OR (lasers[Title/Abstract])). The relevant articles were entered into the Science Citation Index to retrieve reports citing them, and the reference list of all the articles obtained were filtered.
Inclusion and exclusion criteria
The RCTs, cohort trials, and retrospective studies aiming to compare laser techniques versus mechanical staplers in pulmonary resections were included in our review. The surgical approach was thoracoscopy. Bilateral and unilateral procedures were considered, and only English articles were included for systematic review. We excluded studies if they exclusively focused on nonoperative management or nonthoracoscopic procedures or were performed using a less common approach: needlescopic approaches or single-incision approaches.
Selection of studies
The title and abstract of literature considered to be potentially relevant were independently filtered by 2 reviewers (XS Liu and PL Zhang). After retrieving the full text of these articles, we filtered the studies using the inclusion and exclusion criteria. We resolved differences in study selection by consulting with the third reviewer (Y Zhang). We determined the excluded studies based on the above criteria and stated the reason. When selected studies were included in our research, we could use the following domains to assess their methodological quality and extract and analyze data.
Quality assessment
Two reviewers (Liu and Qian) evaluated the methodology of included literatures independently based on random sequence generation, allocation concealment, blinding of participants and personnel, blinding of outcome assessment, incomplete outcome data, and selective reporting. Each participant had an equal chance of being allocated to a group by randomized method, indicating that the risk of bias in random allocation was low. If the investigators in charge of patient selection could not determine the next treatment in allocation (e.g., central randomization; sealed, opaque, sequentially numbered assignment envelopes), the risk of bias in allocation concealment was considered low. If the patients were blind to the treatment assignment, the risk of bias of blinding of participants and personnel was considered low. If participants of the outcome assignment were blind to group assignment, the risk of bias of blinding of outcome assessors was considered low. The risk of incomplete outcome data caused by the use of intent-to-treat analysis in research was considered low. If researchers provided a prespecified published protocol available for comparison, the risk of bias of selective reporting was considered low. We categorized each item of bias as a low risk, a high risk, or unclear and present this information in Supplemental Table 1. Disagreements in the bias assessment were resolved through discussions with a third expert (Y Zhang).
Outcome measures
The prespecified outcomes of our qualitative meta-analysis were as follows: efficacy outcomes (operating time, hospital stay, and re-operation) and safety outcomes (mortality, prolonged air leak, pneumothorax, and other complications). Disability and health status, lung function, and analysis of costs were also included in our analysis.
Data extraction
XS Liu extracted data on author, year of publication, patient characteristics (age, gender, body mass index, history of smoking, and lung function), study design, criteria for the study, and clinical outcome from the trials using a standardized form (Tables 1 and 2) and major clinical outcomes for qualitative analysis. PL Zhang or Y Zhang independently cross-checked the extracted data. Any disagreement was resolved through discussion.
Statistical analysis
RevMan 4.0.4 (The Cochrane Collaboration, Wintertree Software Inc, Canada) was used to summarize relative risk (RR) and weight mean difference (WMD), which were expressed as point estimates in square brackets with a 95% confidence interval (CI). Because only 3 articles were included in our research, we performed a qualitative systematic review without summarizing the results. When 1 was included in the 95% CI of RR, there was no difference in outcomes between the 2 groups. When 0 was included in the 95% CI of the WMD, the difference between the 2 groups was considered not significant. We transformed median and range into mean and standard deviation according to the method of Hozo if necessary.8
Results
Inclusion
The initial electronic searches identified titles from which abstracts were obtained to identify 799 potentially relevant studies of pulmonary resection comparing laser techniques with mechanical staplers from 1975 to 2022. After initially removing duplicated articles and screening titles and abstracts, 22 publications were included for full-text search. Due to the reasons below, 19 articles were excluded: lack of comparative trials (n = 10), lack of laser techniques (n = 4), lack of mechanical staplers (n = 3), and low quality (n = 2). Only 3 studies appeared to meet the criteria for this review to qualitative analysis (Fig. 1). Two reviewers (XS Liu, K Qian) independently assessed the searching results and resolved the disagreements through discussion. The flow chart of the searching process is given as shown in Fig. 1.
Citation: International Surgery 109, 3; 10.9738/INTSURG-D-22-00014.1
Author verification
The author of this study submitted the data extraction sheets to verify the accuracy of information obtained from the paper and to provide other relevant information if necessary. But no reply was received.
Assessment of quality
The risk of random sequence generation was low in 1 trial of Scanagatta and unclear in the other 2 trials. For all trials, the risks of allocation concealment, blinding of outcome assessment, and selective reporting were considered unclear, whereas the risk of blinding of participants was considered high due to the nature of surgery procedure. The risk of incomplete outcome data was unclear in 2 trials and low in the trial of Scanagatta due to the intent-to-treat analysis (Figs. 2 and 3 and Supplemental Table 1).
Citation: International Surgery 109, 3; 10.9738/INTSURG-D-22-00014.1
Citation: International Surgery 109, 3; 10.9738/INTSURG-D-22-00014.1
Characteristics of participants
The participants were divided into the laser group and the staple group in all included studies. Participants in the study of McKenna had diffuse emphysema treated by bullectomy using the video-assisted thoracoscopic surgery (VATS) approach.9 Participants in the study of Marulli had lung cancer scheduled for elective pulmonary lobectomy using the VATS approach.10 Participants in the remaining study were immunocompetent adults scheduled for anatomic pulmonary resection using the VATS approach.1 Our study included 188 participants in total. The study population consisted mainly of 134 males (71%) with an average age of 67 years (Table 1).
Type of intervention
In the laser group, a variety of types of lasers were used: a contact tip Nd:YAG laser (10 W),9 a noncontact thulium laser (40 W),10 and a photocoagulation-mode thulium laser.1 Scanagatta did not report the laser power output. For the staple group, McKenna et al used a 60-mm endoscopic stapler (ELC 60, Ethicon, Inc, Somerville, NJ) with bovine pericardium (Peristrips, Biovascular, St. Paul, MN) to buttress the staples,9 Marulli et al applied standard staplers to complete pulmonary fissures,10 and Scanagatta completed the division of fissures using staplers with sealants when needed1 (Table 2).
Efficacy Outcomes
Operating time
In general, the operating time is defined as the time from skin incision to skin closure. In both studies, the laser group and the staple group did not have significant differences in the operating time (WMD = 12.00 minutes, 95% CI −11.66 to 35.66, and WMD = 2.00 minutes, 95% CI −15.10 to 19.10). However, Marulli reported the mechanical staplers apparently decreased the operating time compared with the lasers (WMD = 39.00 minutes, 95% CI 21.82 to 56.18) (Fig. 4A).
Citation: International Surgery 109, 3; 10.9738/INTSURG-D-22-00014.1
Hospital stay
In the 3 trials, there was no significant difference in mean difference between the 2 instruments (WMD = −2.00 days, 95% CI −7.36 to 3.36 for McKenna; WMD = −3.00 days, 95% CI −6.29 to 0.29 for Marulli; and WMD = 0.00 days, 95% CI −1.69 to 1.69 for Scanagatta) (Fig. 4B).
Re-operation
RR (95% CI), expressed as the re-operation rate of the laser group versus the staple group, was RR = 1.18, 95% CI 0.08 to 18.17 for McKenna and RR = 2.69, 95% CI 0.11 to 63.96 for Scanagatta. Marulli reported that no patients required re-operation in both groups (Fig. 4C).
Adverse Events
Mortality
McKenna reported no perioperative deaths in the laser group and 1 death caused by contralateral tension pneumothorax in the staple group. No perioperative mortality was observed in both groups of Marulli and Scanagatta.
Persistent air leaks
The persistent air leaks were identified when air leaks lasted more than 7 days.11 There was no significant difference in the incidence of persistent air leaks between the laser group and the staple group in the 3 studies (RR = 0.68, 95% CI 0.38 to 1.22; RR = 0.67, 95% CI 0.12 to 3.61; and RR = 1.07, 95% CI 0.53 to 2.16) (Fig. 4D).
Pneumothorax
McKenna reported that 6 patients developed delayed pneumothorax in the laser group, and 1 developed tension pneumothorax in the contralateral lung in the staple group (RR = 7.09, 95% CI 0.90 to 55.95). Marulli reported pneumothorax in only 1 patient in the staple group (RR = 0.33, 95% CI 0.01 to 7.76). Scanagatta reported 3 cases of delayed pneumothorax in the laser group (RR = 6.28, 95% CI 0.34 to 117.39) (Fig. 4E).
Other complications
Our review indicated that there was no difference in the incidences of other complications, including fever, bleeding, pneumonia, chylothorax, respiratory failure, atrial fibrillation, deep vein thrombosis, transient cerebral ischemia, and ileus in both groups (RR = 0.16, 95% CI 0.01 to 3.13 for McKenna; RR = 0.42, 95% CI 0.09 to 1.96 for Marulli; and RR = 1.11, 95% CI 0.41 to 3.03 for Scanagatta) (Fig. 4F).
Disability and health status
McKenna reported breathlessness probably measured using the Medical Research Council or Rand Dyspnea scale, showing that 8/33 (24%) patients in the laser group had improved dyspnea by more than 1 grade compared with 26/39 (66%) patients in the staple group (p ＝ 0.003). Marulli and Scanagatta did not report this information in their studies.
Lung function
McKenna reported a mean improvement in forced expiratory volume in 1 second at 6 months of 13.4% ± 5.5% and 32.9% ± 4.8% in the laser group and the staple group, respectively (p < 0.01). The increase in forced vital capacity was similar: 6% ± 3% (the laser group) versus 21% ± 6% (the staple group) (p = 0.07). Marulli and Scanagatta did not report changes of lung function after pulmonary surgery.
Cost analysis
Marulli found the hospitalization cost in the stapler group was significantly higher than that in the laser group: 5650 ± 3063 euros for the laser group versus 8147 ± 5785 euros for the staple group (P = 0.01). Additionally, an equally significant result of Scanagatta reported the intraoperative cost of the 2 procedures: 807 ± 212 euros for the laser group versus 1092 ± 274 euros for the staple group (P < 0.0001).
Discussion
The efficacy and safety of lasers and mechanical staplers for pulmonary surgery were compared through 3 RCT trials identified by systematically searching the literatures. Currently, the evaluation of various standard techniques widely applied in pulmonary resection involving surgical stapling, electrocautery, and hand sewing and their possible roles in thoracic surgical operations as surgical tools was performed. The effect of the Nd:YAG laser on pulmonary parenchyma was first demonstrated by Minton in 1967, when it was used to precisely excise the tumor in an experimental model.12 In the next few years, a new generation laser system, the 2010-nm wavelength emitted by Cyber TM thulium laser, was introduced into the field of thoracosurgery. Since the 1980s, the application of lasers in resections of pulmonary tissue has been widely accepted, and the effect on parenchyma-sparing surgery of various types of lasers was evaluated.13 To our knowledge, there were no systematic reviews to compare the 2 procedures.
Considering the efficacy of the 2 methods, there was no significant difference in the operating time between the laser group and the staple group in 2 trials, whereas Marulli demonstrated that the use of mechanical staplers significantly decreased the operating time compared with lasers, which may be due to their application of a second thulium laser irradiation at low power (20 W) in a defocused mode treating the dissected surface of the residual lobe so as to achieve the ideal aero-hemostatic effect on lung parenchyma. In our systematic reviews, the overall hospital stay was shorter in the laser group, though without statistical significance, possibly due to the limited patients enrolled. Three RCTs regarding the occurrence of re-operation showed no difference in the re-operation rate between the 2 groups. In our clinical experience, the lasers had the same efficacy as the mechanical staplers.
Factors influencing postoperative persistent air leaks include underlying lung diseases, such as emphysema, fibrosis, tuberculosis or malignancies, lymphangioleiomyomatosis, and intrathoracic adhesions, elderly patient (>75 years), and lower diffusion capacity for carbon monoxide.14 In the case of pulmonary traumatization and dissection, persistent air leaks cause prolonged intercostal drainage, associated pain, increased immobility, and risk of further complications such as pneumonia, empyema, and pulmonary embolism.15 Although many efforts have been made to reduce the occurrence of parenchymal air leaks after pneumonectomy, the desirable surgical techniques or tools to reduce or prevent this complication have not been determined, and in most cases, standard techniques cannot provide adequate sealing in patients. Mineo et al reported 3 positive effects of laser irradiation (cutting, coagulation, and tissue shrinkage), leading to sealing through the progressive collapse of alveolar septa, which produces a thick and multilayer air-proof membrane.16 But we observed no significant difference in the incidence of persistent air leaks between the 2 groups, possibly because the sample sizes in these studies were small. However, the laser group showed a daily trend of reduction in the proportion of patients with persistent air leaks.
Additionally, pneumothorax and other complications, such as fever, bleeding, pneumonia, chylothorax, respiratory failure, atrial fibrillation, deep vein thrombosis, transient cerebral ischemia, and ileus were distributed equally in each group. Considering the high blood vessel density, it is necessary to remove the lung parenchyma with a laser with a strong coagulation ability and excellent cutting performance. With respect to the mortality outcome, the results of the 3 RCTs included concluded that the outcome of mortality was rare (<5% staple group event rate). These results can be explained by the features of the laser, which can be strongly absorbed by water, resulting in excellent coagulation (low coagulation depth of only 0.2 mm) and aero-hemostatic effects, and maximal preservation of surrounding normal tissues.
Rogers and colleagues found that removal of large bullae in isolated bullous disease and bullous emphysema may lead to improved expiratory flow and airway conductance, which may be due to the improved elastic recoil.17,18 In 1996, an RCT containing 2 procedures (laser bullectomy and lung reduction surgery with staples) was conducted using the VATS method, and the results strongly support the superiority of the unilateral staple procedures over laser methods in terms of improved quality of life, pulmonary function, and probably reduced oxygen dependence.9 Although the studies of Marulli and Scanagatta did not report these outcomes, they applied a new generation of laser (a Cyber TM thulium surgical laser system), and this may result in a better clinical status and pulmonary function in the laser group.1,10
Above all, we summarized some advantages of laser mechanisms in pulmonary resections: (1) the minimal surrounding tissue damage and maximal preservation of normal parenchyma elasticity, (2) the aero-hemostatic effect, (3) the relative safety of major bronchi and vessels during the dissection, and (4) the system’s cost-effectiveness. Collectively, we concluded that lasers may be an efficacious and safe alternative to the standard stapling technique in terms of operating time, hospital stay, re-operation, mortality, persistent air leaks, pneumothorax, postoperational complications, and cost.
The studies in this systematic review include many biases in terms of random sequence generation, allocation concealment, blinding of participants and personnel, blinding of outcome assessment, incomplete outcome data, and selective reporting. Long-term health outcomes are unclear. It is hoped that further RCTs will be necessary to better define these issues establishing the advantages of the laser techniques over mechanical staplers.
PRISMA flowchart for literature searching.
A risk of bias graph.
A risk of bias summary (“+” low risk; “?” unclear risk; “-” high risk).
Forest plot of operation time (A), hospital stay (B), re-operation (C), persistent air leaks (D), pneumothorax (E), other complications (F).
Contributor Notes