Quasilobar minimalist lung volume reduction surgery

Abstract

 

Open in new tabDownload slide

OBJECTIVES

Our goal was to assess the results and the costs of the quasilobar minimalist (QLM) thoracoscopic lung volume reduction (LVR) surgical method developed to minimize the trauma from the operation and the anaesthesia and to maximize the effect of the lobar volume reduction.

METHODS

Forty patients with severe emphysema underwent QLM-LVR that entailed adoption of sole intercostal block analgesia and lobar plication through a single thoracoscopic incision. Results were compared after propensity matching with 2 control groups undergoing non-awake resectional LVR with double-lumen tracheal intubation or awake non-resectional LVR by plication with thoracic epidural anaesthesia. As a result, we had 3 matched groups of 30 patients each.

RESULTS

Baseline forced expiratory volume in 1 s, residual volume, the 6-min walking test and the modified Medical Research Council dyspnoea index were 0.77 ± 0.18, 4.97 ± 0.6, 328 ± 65 and 3.3 ± 0.7, respectively, with no intergroup difference after propensity score matching. The visual pain score was better (P < 0.007), the hospital stay was shorter (P < 0.04) and overall costs were lower (P < 0.04) in the QLM-LVR group than in the control groups. The morbidity rate was lower with QLM-LVR than with non-awake resectional-LVR (P = 0.006). Significant improvements (P < 0.001) occurred in all study groups during the follow-up period. At 24 months, improvements in residual volume and dyspnoea index were significantly better with QLM-LVR (P < 0.04).

CONCLUSIONS

QLM-LVR proved safe and showed better perioperative outcomes and lower procedure-related costs than the control groups. Similar clinical benefit occurred at 12 months, but absolute improvements in residual volume and dyspnoea index were better in the QLM-LVR group at 24 months.

INTRODUCTION

Lung volume reduction (LVR) surgery carried out by non-anatomical resection of destroyed areas of the lung [1, 2] continues to prove highly effective as a palliative surgical treatment in selected patients with severe emphysema [3, 4]. Nonetheless, so far, this therapy remains largely underused, and the meaningful 7.9% mortality and 58.7% morbidity have been previously reported in a large study of this treatment modality [5].

To reduce morbidity while assuring equivalent clinical benefit, a number of alternative bronchoscopic LVR methods have been proposed, although so far each has shown advantages and limitations in the clinical setting [6–8].

Among the evolutionary interventional treatment options, a surgical non-resectional LVR method with the patient under awake anaesthesia that we had previously proposed [9–11] has shown promise in offering clinical improvements that paralleled those of the surgical resectional LVR, with lower morbidity, possibly due to the avoidance of general anaesthesia with double-lumen tube intubation and mechanical ventilation and to the avoidance of lung tissue resection.

More recently, taking into account our previous experience as well as the currently investigated bronchoscopic lobar LVR by valve placement, which can induce in selected patients complete atelectasis of the targeted lung lobe [7], we have developed a novel quasilobar minimalist (QLM) LVR method aimed at achieving a nearly complete plication of the targeted lung lobe through a single thoracoscopic access route and an anaesthesia protocol entailing an intercostal block with target control sedation and maintenance of spontaneous ventilation.

The theoretical advantages of this novel surgical method rely on the belief that it might permit us to pursue a safe and more rapid perioperative recovery for the patients who are operated on. We also reasoned that, due to the typically non-anatomical distribution of emphysema, QLM-LVR achieved with no resection and with maintenance of viable bronchi would allow a maximized volume reduction effect without the risks of removal of functional lung tissue and with no need of endobronchial devices.

The goal of this study was to present comparative results of QLM-LVR versus those of 2 control groups undergoing either the standard non-awake resectional (NAR) or the awake non-resectional (NAR) LVR methods.

MATERIALS AND METHODS

Forty patients undergoing unilateral QLM-LVR between January 2016 and December 2019 were included in the study, which was approved by the Tor Vergata ethical committee. All patients provided written informed consent.

The preoperative workup included spirometry with plethysmography to assess respiratory function including forced expiratory volume in 1 s (FEV₁), forced vital capacity (FVC), residual volume (RV) and total lung capacity. Arterial oxygen tension (PaO₂) and arterial carbon dioxide tension (PaCO₂) were assessed by blood gas analysis. Exercise capacity was assessed by the 6-min walking test (SMWT). Dyspnoea was determined by the modified Medical Research Council dyspnoea index (DI). The physical functioning (PF) quality-of-life domain score was assessed by the short-form 36-item questionnaire (SF-36). Radiological morphology of emphysema was evaluated by high-resolution computed tomography of the chest.

The study was designed as a retrospective analysis of a prospective database in which all patient data have been stored according to a previously established standardized protocol.

Results achieved in the QLM group were compared with those of 2 historical control groups undergoing either standard NAR-LVR with general anaesthesia with double lumen tube intubation and mechanical ventilation or ANR-LVR under thoracic epidural anaesthesia, before the current study period.

The surgical strategy has been that of a staged bilateral treatment with initial unilateral LVR in the most severely diseased lung according to semiquantitative analysis of radiological morphology of emphysema [12] and the accomplishment of contralateral LVR tailored on the basis of the clinical/functional loss of postoperative improvements.

For the purposes of this study, only the data related to the results of the first surgical procedure have been taken into account, and the occurrence of contralateral LVR has been considered as an end-event like mortality.

To enhance the reliability of the comparisons, a 1:1:1 propensity score matching analysis was carried out to obtain 3 groups of 30 patients each. Eligibility criteria for LVR were the same for the 3 groups and included the finding of severe emphysema with evidence on high-resolution computed tomography scans of upper-lobe predominant disease associated with severe disability despite maximized medical care with an RV >180% predicted (Supplementary Material, Table 1S) [10].

Exclusion criteria included predominant giant bullous disease, homogeneous distribution of emphysema or signs of extensive pleural adhesions on high-resolution computed tomography and diffusion capacity of the lung for carbon monoxide <20% of that predicted. The primary outcome measures were hospital stay and absolute improvement in RV at 12 months. Secondary outcomes included postoperative (24 h) assessment of thoracic pain according to a 10-score visual analogue scale assessed 24 h after surgery, operating room times including anaesthesia time, operative time, weaning time, recovery room time and global in-operating room time (given by the sum of the previously mentioned times); ratio of PaO₂ to fraction of inspired oxygen (PaO₂/FIO₂), and PaCO₂, assessed at 3 fixed time points (preoperative in lateral decubitus, at the end of the procedure and 1 h after completion of the operation). Clinical improvements including changes in the DI, FEV₁, FVC, PaO₂, SMWT and SF-36 PF were assessed postoperatively at 12 and 24 months.

In addition, we computed procedure-related costs including anaesthesia and surgical devices, laboratory assessment, dressing materials, medications and operative time and hospital-stay-related costs.

Criteria for discharge were standardized and included a stable clinical condition with no complications requiring in-hospital care. Criteria for chest tube removal included 24-h drainage ≤200 ml with no air leaks. Patients with minor air leaks not interfering with lung expansion without suction were allowed to be discharged with a tube connected to a Heimlich valve.

Operative complications (morbidity) were considered to be all adverse events occurring during the first 30 days after surgery and requiring additional medical and/or surgical care.

Anaesthesia and surgical technique

In group QLM, target control sedation was achieved by bispectral index monitoring and intravenous infusion of propofol and remifentanil or dexmedetomidine. A supraglottic laryngeal mask was inserted, and the patient was kept in spontaneous ventilation with by-necessity manually assisted ventilation to keep oxygen saturation > 90% and end-tidal carbon dioxide tension ≤50 mmHg [13].

All surgical procedures were performed by video-assisted thoracoscopic surgery. In group QLM, multiple plications of the upper lobe were performed by first creating flaps of lung tissue anteriorly along the fissure and posteriorly, as deep as possible, by non-cutting staple suturing. Subsequently, a second line of sutures was fired after we grasped together the anterior and posterior flaps in order to reduce the volume of the upper lobe by about 80%. One 15 Fr transcutaneous thoracic catheter (REDAX S.P.A. Poggio Rusco (MN), Italy) was inserted at the end of the procedure (Figs 1 and 2 and Video 1). Differences in anaesthesia and surgical techniques among the study groups are detailed in Table 1.

Statistical analyses

Statistical analyses were performed using the IBM SPSS Statistical Package (IBM SPSS Statistics for Windows, Vers. 25.0, 2017; Armonk, NY, USA) and app R version 2.12.1(R Foundation for Statistical Computing, Institute for Statistics and Mathematics, Vienna, Austria).

Descriptive statistics consisted of the mean ± the standard deviation for parameters with Gaussian distributions (after confirmation with histograms and the Kolmogorov–Smirnov test).

Following a crude analysis of the entire cohort, a 1:1 propensity score matching was performed to create 2 homogeneous cohorts: 1 investigational group included 30 QLM patients versus 1 control group including 30 patients who underwent NAR plus 30 patients who underwent ANR, considering as covariates age, body mass index (BMI), FEV₁ and RV. For each group, the standardized mean difference with range and 95% confidence interval was calculated; subsequently, the homogeneity between the NAR and ANR groups was tested by one-way analysis of variance using the same covariates age, BMI, FEV₁ and RV.

Comparisons among the 3 study groups, QLM, NAR and ANR, were performed with the analysis of variance one-way and post hoc using the Bonferroni test or analysis of variance for repeated measures with comparisons using the Bonferroni test and using the χ² test or the Fisher’s exact test (if cells <5) for frequencies.

Survival was assessed using the Kaplan–Meier method, and the log-rank test was used to match curves. A P-value <0.05 was considered statistically significant. Outcome analysis was carried out on an intention-to-treat basis.

RESULTS

Baseline data of the prematching cohorts showed a significant intergroup difference in BMI and FEV₁ among groups NAR versus ANR; in RV, total lung capacity and PaCO₂ between groups QLM versus NAR and in the SF-36 PF domain score between groups QLM and ANR (Supplementary Material, Table 2S).

Following propensity score matching, intergroup differences were no longer detected in age, BMI, FEV₁ and RV (Fig. 3).

Analysis of operative times has shown that in the QLM group, anaesthesia time, operative time and global time spent in the operating room were shorter than in both control groups (Table 2).

Changes in oxygenation and PaCO₂ at fixed time points during surgery are depicted in Fig. 4, which shows that significant perioperative changes occurred over time with better recovery towards preoperative status 1 h after surgery in groups QLM and ANR.

There was no conversion to thoracotomy in any group nor was there conversion to general anaesthesia and intubation in the QLM and NAR groups.

There were no operative deaths. The postoperative pain visual analogue scale was lower in the QLM group whereas morbidity was similar between groups QLM and ANR and higher in the NAR group (Table 2). In particular, prolonged air leaks (>7 days) occurred in 4, 6 and 16 patients in the QLM, ANR and NAR groups, respectively. Other complications included atrial fibrillation in 3 patients (2 in the NAR group and 1 in the ANR group), pneumonia in 1 patient in the NAR group; and wound infection requiring debridement and surgical resuturing in 1 patient in the QLM group.

Hospital stay was significantly shorter in the QLM group than in both control groups (Table 2). In particular a hospital stay ≤5 days occurred in 3 patients in group NAR (range 4–24 days), in 11 patients in group ANR (range 3–34 days) and in 18 patients in group QLM (range 2–14 days). Procedure-related costs were lower in the QLM group than in both control groups (Table 2).

Postoperatively, significant clinical improvements occurred at 12 months in RV, FEV₁, FVC, SMWT distance, DI and the SF-36 PF domain score in all study groups whereas at 24 months absolute improvements in RV and DI were greater in the QLM group (Table 3).

During the follow-up period, no patient in any study group was readmitted due to pulmonary infections in the operated lung. Survival and freedom from contralateral treatment are depicted in Fig. 5. Overall, 5 patients died within 12 months of respiratory failure, 1 in the QLM group, 1 in the ANR group and 3 in the NAR group. At 24 months, there was no inter-group (QLM vs ANR vs NAR) difference in survival (95% vs 87% vs 87%) or in freedom from contralateral treatment (92% vs 89% vs 85%).

DISCUSSION

QLM-LVR proved safe and feasible and resulted in shorter hospital stays and in lower procedure-related costs than in the control groups. Postoperatively, significant improvements occurred at 12 and 24 months in RV as well as in other respiratory function and clinical outcome measures in all study groups. Furthermore, 24-month improvements in RV and DI were greater in the QLM group.

According to a so-called minimalist surgical strategy [13], we adopted a single operative port and a single, small chest drain, which are likely to explain the lower pain visual analogue scale at 24 h observed in the QLM group.

The QLM-LVR method entailed a change from fully awake anaesthesia through thoracic epidural anaesthesia to sole intercostal block analgesia with target control sedation and maintenance of spontaneous ventilation, which eliminated the discomfort occurring when operating on conscious subjects. Our protocol included the placement of a laryngeal mask, which allowed us to easily correct permissive hypercapnia, which may occur in video-assisted thoracoscopic surgery LVR under spontaneous ventilation [11]. The reliability of our anaesthesia protocol is supported by the absence of conversions to general anaesthesia and tracheal intubation in the QLM group. Moreover, maintenance of spontaneous ventilation led to improvement in the immediate postoperative behaviour of oxygenation and ventilation in patients who were operated on, resulting in a rapid return of PaO₂/fraction of inspired oxygen and of PaCO₂ towards preoperative values within 1 h after surgery (Fig. 4) [9–11]. As a result, global operating room time was significantly shorter in the QLM group, particularly when compared with the NAR-LVR group. Our results also compare favourably with those of the National Emphysema Treatment Trial research group [5] in which reintubation, inability to wean from ventilator and discharge to home occurred in 21.8%, 8% and 69.7%, respectively.

In our study groups, there were no operative deaths. This result enhances another recent study [4] finding of a mortality rate of 2.2% in a series of 135 patients with similar characteristics to those recruited in the National Emphysema Treatment Trial research group [5] in which a 90-day mortality of 7.9% was reported. These findings suggest an ongoing improvement both in the surgical techniques and clinical management of these delicate patients in experienced centres [4].

The satisfactory results achieved by QLM-LVR enhance previous findings that ANR-LVR under spontaneous ventilation can offer highly satisfactory clinical improvements with lower morbidity and shorter hospital stay than the standard NAR-LVR method [11]. The low morbidity rate observed in the QLM-LVR group must be mainly attributed to the reduction in prolonged air leaks compared to the NAR-LVR group. In fact, air leaks represent the most frequent adverse event after LVR with resection occurring in up to 90% of patients postoperatively and lasting more than 30 days in about 12% of instances [14]. In our study, prolonged air leaks (>7 days) occurred in the QLM- and ANR-LVR groups in 13% and 20% of patients, respectively, versus in 53% of patients in the NAR-LVR group. Though reasons underlying this finding remain speculative, we believe that with QLM-LVR there is no pleural discontinuation and the multiline lobar plication is less likely to develop air leaks.

The cost-effectiveness of LVR has been questioned in previous studies [15]. In the current analysis, procedure-related costs were lower in the QLM group compared to those in both control groups, mainly due to shorter operating room and hospitalization times, which overpassed in cost saving the slightly superior cost due to the greater number of cartridges needed on average for the QLM-LVR group (Table 1).

In our study groups, significant improvements occurred in RV, FEV₁, FVC, SMWT and SF-36 PF quality of life domain at 12 and 24 months after surgery, although at 24 months, absolute improvements in RV and DI were significantly greater in the QLM group.

The significant magnitude of improvements we achieved by unilateral LVR in all study groups confirms the importance of selection criteria that included in our series patients with upper-lobe predominant emphysema with limited exercise capacity. In fact, a similar cohort has already been reported to gain significant improvement in mortality as well as in exercise capacity and symptoms for up to 5 years following one-stage bilateral treatment [16, 17]. On the other hand, significant benefits of LVR have also been reported in patients with homogeneous distribution of disease [18] and, more recently, in patients with lower-lobe predominant disease [19]. In this regard, the potential benefit of QLM-LVR in these subcohorts is under evaluation at our centre.

In addition, other potential technical advances, including robotic LVR with firefly fluorescence perfusion to potentially improve identification of tissue to resect or plicate, is currently being explored clinically [20].

Our study results are referred to the outcomes of unilateral treatment within a staged bilateral strategy that we have preferred since 2002 [21]. The primary reason for our choice is to spare our patients who were operated on the discomfort of a simultaneous bilateral LVR. Moreover, we previously reported [21] greater absolute improvements in RV, FEV₁ and FVC following one-stage bilateral LVR but more stable long-term benefits following staged bilateral procedures.

Limitations

This study has some limitations. First, it entails a retrospective non-randomized investigation, which includes the typical risks of bias. These risks are hopefully mitigated by the inclusion of a propensity score matching, which permitted increased homogeneity and comparability of the study cohorts. In addition, a learning curve effect might have affected operative time for the QLM-LVR procedure, although we attribute it to the maximized simplification of our novel surgical technique. Finally, we acknowledge that the choice of RV as a primary outcome might be considered an uncommon metric for this purpose. However, the main proposed mechanisms underlying post-LVR improvements primarily involve a reduction in hyperinflation leading to restored chest wall and diaphragm mechanics, restored lung elastic recoil and radial traction, eventually resulting in improved expiratory flow rates and lung emptying [22]. Thus we considered it rational to assume that LVR should have as a primary goal a significant and stable reduction in RV.

CONCLUSION

QLM-LVR resulted in shorter hospital stays, lower procedure-related costs, similar significant clinical improvements at 12 months and superior benefit in RV and DI at 24 months compared to the control groups undergoing either NAR-LVR under general anaesthesia with tracheal intubation and one-lung ventilation or ANR-LVR under thoracic epidural anaesthesia. We hypothesize an increasing role of QLM-LVR in the management of selected patients with severe emphysema. Further controlled studies are warranted to confirm our encouraging findings.

SUPPLEMENTARY MATERIAL

Supplementary material is available at EJCTS online.

Presented at the European Society of Thoracic Surgeons Annual Meeting 2020.

ACKNOWLEDGEMENTS

This study was performed within a collaborative clinical research programme established between the University Tor Vergata of Rome, Italy, the Tel Aviv Medical Center, Israel and the Tanta University, Egypt.

Conflict of interest: none declared.

Author contributions

Eugenio Pompeo: Conceptualization; Data curation; Formal analysis; Investigation; Methodology; Supervision; Validation; Visualization; Writing—original draft; Writing—review & editing. Ahmed Elkhouly: Data curation; Formal analysis; Supervision; Writing—original draft; Writing—review & editing. Paola Rogliani: Data curation; Formal analysis; Resources; Supervision. Mario Dauri: Conceptualization; Data curation; Resources; Supervision; Writing—review & editing. Michael Peer: Data curation; Investigation; Resources; Supervision. Gianluigi Sergiacomi: Investigation; Methodology. Roberto Sorge: Data curation; Formal analysis; Methodology; Software; Supervision; Validation.

Reviewer information

Reviewer information European Journal of Cardio-Thoracic Surgery thanks Michael Rolf Mueller, Karen C Redmond, Mohamed Salama and the other, anonymous reviewer(s) for their contribution to the peer review process of this article.

REFERENCES

ABBREVIATIONS

 
• ANR

  Awake non-resectional

 
• BMI

  Body mass index

 
• DI

  Dyspnoea index

 
• FEV₁

  Forced expiratory volume in 1 s

 
• FVC

  Forced vital capacity

 
• LVR

  Lung volume reduction

 
• NAR

  Non-awake resectional

 
• PaCO₂

  Arterial carbon dioxide tension

 
• PaO₂

  Arterial oxygen tension

 
• PF

  Physical functioning domain

 
• QLM

  Quasilobar minimalist

 
• RV

  Residual volume

 
• SF-36

  Short-form 36-item

 
• SMWT

  6-min walking test