Anterolateral vs. Clamshell Thoracotomy for Bilateral Lung Transplantation

Authors: Sanne J. J. Langmuur, Leonard Seghers, Rogier A. S. Hoek, Jos A. Bekkers, Maarten ter Horst, Edris A. F. Mahtab

Abstract

Background

Clamshell thoracotomy is the traditional approach for bilateral sequential lung transplantation (LuTx), despite considerable morbidity. The bilateral anterolateral thoracotomy without sternal division is proposed as an alternative, less invasive method. Since this technique is more surgically challenging, the aim of this study was to evaluate the more favorable surgical approach.

Methods

Adult patients undergoing LuTx between 2010 and 2022 were included in this single‐center retrospective cohort study. Multivariable regression analyses were performed to assess the effect of surgical approach on mortality, operative outcomes, and complications.

Results

A total of 249 patients (anterolateral n = 132, n = 117) were included. Recipients in the anterolateral thoracotomy group were older, on the waiting list for a shorter period of time, classified less often as highly urgent, and differed in their indications for LuTx. After multivariable correction, operating and ischemic times were longer for patients in the anterolateral thoracotomy group. However, patients with an anterolateral thoracotomy incision had less intraoperative extracorporeal circulation (ECC) use, blood loss, and need for transfusion. While survival was similar, ICU and hospital stay, and duration of mechanical ventilation were shorter. Patients in the anterolateral thoracotomy group also had significantly less wound‐related complications. The use of intraoperative ECC, which was much higher in the clamshell group, seemed to play an important role in these differences in outcomes.

Conclusion

Performing LuTx through the less invasive anterolateral thoracotomy is a favorable alternative to the clamshell incision. Despite increased technical difficulty and a limited increase in operating and ischemic times, survival is similar, but the anterolateral thoracotomy patients had significantly less peri‐ and post‐operative complications and a faster recovery.

Introduction

Traditionally, the routine surgical approach for bilateral sequential lung transplantation (LuTx) is to perform a bilateral transverse sternothoracotomy, also known as a clamshell incision [1]. This incision is used for various procedures in cardiothoracic and trauma surgery, but is known for being very invasive, and considerable morbidity is related to this incision [2, 3, 4]. Therefore, the bilateral anterolateral thoracotomy without sternal division is proposed as an alternative, less invasive approach. This technique has been demonstrated to result in less postoperative trauma, shorter intensive care unit (ICU) stay, and duration of mechanical ventilation than the clamshell incision, whilst survival remains the same [3, 5, 6, 7]. Moreover, studies showed lower blood loss and transfusion need, less wound‐related complications, and less postoperative lung restriction [3, 4, 5, 7, 8]. However, some surgeons are still reluctant to use this less invasive technique because it is more surgically challenging, possibly causing ischemic times to be longer [5]. Also, the anterolateral thoracotomy has been associated with higher pain scores than the clamshell incision [6].

Largely due to complications after the clamshell incision, the preferred approach for LuTx in our center changed to performing a bilateral anterolateral thoracotomy. Although some prior studies have investigated the effect of the incision on patient outcomes, the majority did not correct for possible confounding, despite mostly using a historical cohort as a control group. Therefore, the aim of this study was to investigate whether patient recovery is enhanced for LuTx recipients operated through anterolateral versus clamshell thoracotomy, by performing an in‐depth comparison of operative characteristics and short‐term outcomes in one of the largest single‐center cohort studies to date, whilst adjusting for possible confounders.

Methods

This is a single‐center retrospective cohort study including adult patients undergoing bilateral LuTx in the Erasmus Medical Center, Rotterdam, between January 2010 and December 2022. The study complies with the International Society for Heart and Lung Trans‐plantation (ISHLT) Ethics Statement. Written informed consent for use of anonymous data was obtained from all patients prior to transplantation (MEC‐2019‐0291, date of approval 14‐07‐2022). Patients with re‐LuTx were excluded from the study. Patients were operated via a clamshell incision (clamshell group) or a bilateral anterolateral thoracotomy (anterolateral thoracotomy group). When LuTx was initially started with an anterolateral thoracotomy incision but had to be converted to a clamshell incision during the procedure, the patient was ultimately included in the clamshell group. If conversion to a clamshell incision occurred during a rethoracotomy in the same admission, the patient was excluded from analysis.

Data Collection

All patient data was collected from the electronic medical records. Data on operating and ischemic times, extracorporeal circulatory support (ECC) (extracorporeal membrane oxygenation [ECMO] or cardiopulmonary bypass [CPB]), blood loss, and transfusion were collected from the operation registration. Total operation duration was calculated from the time of the first incision to the closure of the skin. Implant times were calculated as from the time of incision to the time of taking the donor lung into circulation in the recipient. Blood loss was registered as the amount of blood collected by the cellsaver (fluids other than blood excluded) and drapes and gauzes that were weighed intraoperatively. Transfusion of cellsaver blood was the amount of blood returned from the cellsaver to the patient. Patients were registered to have wound‐related complications in case this was documented in their electronic medical record. Wound complications were considered deep if mediastinitis was present and superficial if they concerned only the soft tissues. Since the lung allocation score (LAS) was introduced in the middle of the inclusion period (2014), patient urgency was determined as highly urgent (HU) or not. Patients who were transplanted after the LAS was introduced were considered HU if they had a LAS ≥50. Before LAS introduction in the Netherlands, patients were granted HU status after a unanimous decision of the national HU committee.

Study Outcomes

Primary outcome measures for this study included 30‐day survival, hospital and ICU length of stay (LOS), and total operation duration. Secondary outcome measures included ischemic and implant times, need for perioperative ECC use, blood loss, perioperative transfusion of cellsaver blood, erythrocytes, platelet concentrate, and plasma volume, mechanical ventilation duration until first detubation, incidence of (hemorrhagic) rethoracotomy during first admission, and wound complications.

Surgical Procedure

Patients underwent bilateral LuTx through either clamshell incision or bilateral anterolateral thoracotomy. The bilateral anterolateral thoracotomy was gradually introduced in our center in 2017. Anesthesiological preparation was performed according to local, standardized protocol. A central venous catheter was introduced via right jugular vein (or in case of probability for veno‐venous ECMO via left jugular vein). An invasive arterial blood pressure catheter was introduced into the radial artery. Patients were intubated with a double‐lumen tube in order to allow for single lung ventilation. Transesophageal ultrasound was used to monitor cardiac function. In case of severe pulmonary hypertension (i.e., in our when the pulmonary arterial systolic pressure was ≥ two‐thirds of the systolic blood pressure), ECMO was installed prior to induction or prior to the start of incision. Intravenous medication such as epoprostenol or treprostinil for treatment of severe pulmonary hypertension was continued during the procedure. Prior to the start of the surgery, the patient was positioned in the supine position with an inflatable cushion under both scapulae in order to selectively lift each hemithorax depending on the operating side, with their arms along their chest and their elbows flexed. In the clamshell group, a submammary incision was performed in the fourth intercostal space. The clamshell incision extended from the left to the right midaxillary line across the anterior aspect of the chest and included transversely splitting the distal one‐third part of the sternum. The internal mammary artery and vein were ligated on both sides. In case of need for ECC, this was initiated through central cannulation in the majority of patients. After transplantation of both lungs, the sternum was approximated with two steel wires, and the incision was closed. The anterolateral thoracotomy incision in LuTx patients was previously described in more detail [9]. In short, two smaller anterolateral incisions were sequentially made in the fourth or fifth intercostal space, extending from the parasternal area to the anterior axillary line. The sternum and mammary vessels were left unharmed. ECC was connected peripherally via the femoral vessels if needed.

Intra‐operative management has largely remained the same over time, except for the introduction of a rotational thrombo‐elastometry (ROTEM) guided algorithm around 2015 to guide transfusion management, whereas this was just clinically‐guided before [10]. Furthermore, a gradual shift from CPB as a method of ECC towards ECMO was observed over the course of this study. Methylprednisolone and mannitol were administered prior to recirculation of each donor lung. A bronchoscopy was performed after lung implantation to evaluate bronchial anastomoses. Postoperatively, potential wound complications were treated by refixating the sternum or ribs in case of dehiscence, and negative‐pressure wound therapy in case of deep wound complications. In case of superficial wound complications, the treatment varied for different sizes and depths of wounds, from daily cleaning and redressing to negative‐pressure wound therapy.

Statistical Analysis

All statistical analyses were performed in R (R Foundation for Statistical Computing, Vienna, Austria, version 4.3.2) and in consultation with a statistician. A two‐sided p value ≤ 0.05 was considered statistically significant. Continuous variables were presented as mean ± standard deviation (SD) if normally distributed, and as median (interquartile range [IQR]) otherwise. Normality was tested using density plots, QQ‐plots, and Shapiro–Wilk tests. Categorical variables were presented as number (%). A Student's t‐test or Mann–Whitney test for continuous variables and the χ ^2^ test for categorical variables were used as appropriate. Multivariable regression was performed to correct for age, sex, body mass index (BMI), diabetes mellitus (DM), HU status, transplant indication, previous thoracic surgery, ex‐vivo lung perfusion (EVLP), and size reduction. In addition, for transfusion outcomes, ROTEM use was added to the model if the p value from the Wald‐test was <0.20 in the full model, as ROTEM had a considerable number of missing values in the transition phase between protocols and only had an effect on plasma volume in LuTx patients in a prior study [10].

Linear regression models were used for continuous outcomes. For LOS and units of erythrocytes and platelet concentrate, Poisson models were used, or quasi‐Poisson in case of over‐ or underdispersion. For binary outcomes, logistic models were used and penalized with ridge regression if necessary. Model assumptions were checked by plotting the residuals. The multicollinearity assumption was assessed by measuring the variance inflation factors, which should have a value ≤5. Stratification for ECC use was performed for the primary outcomes and all blood loss and transfusion parameters, as ECC was regarded as a mediator between the effect of incision and outcome and therefore not included in the multivariable regression analysis.

Results

Between January 2010 and December 2022, 290 patients underwent LuTx in our center. After excluding one patient <18 years, one patient in whom the incision was converted to clamshell during a reoperation, seven patients who underwent retransplantation, and 35 unilateral lung transplants, 249 patients were included for analysis (Figure 1).

Baseline characteristics are presented in Table 1. Patients in the anterolateral thoracotomy group were older, had DM less often prior to transplantation, were on the waiting list shorter and had less high urgency status and hospitalization before transplantation. In addition, indications for LuTx differed between groups, mainly having less cystic fibrosis (CF) and more chronic obstructive pulmonary disease (COPD) in the anterolateral thoracotomy group (Figure 2A).

Operative Characteristics

In total, 132 patients had an anterolateral thoracotomy, and 117 patients underwent a clamshell incision. Of the clamshell patients, 10 (8.5%) initially had an anterolateral thoracotomy incision, which was converted intraoperatively, either strategically or reactively. Reasons for conversion included suboptimal exposure, adhesions, intraoperative bleeding, and hemodynamic instability. The trend over time was illustrated in Figure 2B. Operative characteristics are described in Table 2.

In the anterolateral thoracotomy group, ischemic time was slightly longer for both lungs. The implant time, registered as the time between first incision and end of ischemic time for that lung, was similar for the first lung (176 [148–210] min vs. 182 [153–215] min, p = 0.65), but in favor of the clamshell group for the second lung (302 [267–352] min vs. 337 [298–394], p = 0.002). These differences remained after multivariable regression to correct for confounding (Table 3). Total operation duration did not differ between groups (413 [355–506] min vs. 403 [360–484], p = 0.91), however, it was in favor of the clamshell group after multivariable regression analysis (p = 0.03). All estimates for the multivariable regression analyses can be found in Tables S1 to S16.

Extracorporeal circulatory support (ECMO or CPB) during LuTx was used more often in the clamshell group (75 [64%] vs. 43 [33%], p < 0.001). This consisted mostly of higher CPB use (Table 2). After adjusting, this difference remained statistically significant (OR 0.37 [0.19 – 0.68], p = 0.002). Furthermore, EVLP was more common in the anterolateral thoracotomy group (14% vs. 2%, p < 0.001), and size reduction was less frequent (2% vs. 10%, p = 0.02).

Patients in the clamshell group lost significantly more blood intraoperatively (2918 [1650–5758] mL vs. 1688 [896–3608], p < 0.001). Additionally, they required higher cellsaver blood return, more erythrocytes, platelet concentrate, and plasma volume. When correcting for potential confounders, these differences all remained, except for the difference in plasma transfusion (Table 3).

Early Postoperative Outcomes

A lower number of patients underwent a rethoracotomy after LuTx in the anterolateral thoracotomy group (22 [17%] vs. 39 [33%], p = 0.004). In both groups, the major indication for rethoracotomy was a hemorrhagic complication, such as a rebleed or cardiac tamponade. Other (rare) reasons for rethoracotomy included empyema, a fistula, torsion of a lung, revision of one of the anastomoses, or a chylothorax. For patients needing rethoracotomy, the number of additional surgeries did not differ between groups (1 [1–2] vs. 1 [1–2.5], p = 0.38). However, in the multivariable model, incision type was no longer a statistically significant predictor for rethoracotomy (odds ratio [OR]: 0.55 [95%CI 0.27–1.15], p = 0.11), whereas for example, HU status was (Table S12).

The ICU length of stay was shorter in the anterolateral thoracotomy group (6 [3–23] days) vs. the clamshell group (13 [5–51] days, p < 0.001), as was the hospital LOS (33 [24–55] days vs. 45 [29–92] days, p < 0.001). These differences in favor of the anterolateral thoracotomy group remained after elaborate correction for confounders. The number of days on ventilation after LuTx did not differ significantly between groups (2 [1–19] days vs. 2 [1–5] days, p = 0.07), even though the IQR was much wider in the clamshell group. However, after univariable and multivariable Poisson regression, there was a significant difference in favor of the anterolateral thoracotomy group (incidence rate ratio [IRR]: 0.49 [95%CI 0.26–0.49], p = 0.03). Thirty‐day survival did not differ between groups in unadjusted and adjusted analyses.

Wound‐Related Complications

Wound‐related complications occurred significantly less often in the anterolateral thoracotomy group (p < 0.001) (Table 2). In the clamshell group, a wound complication occurred in 54 (46%) patients. Thirty‐three (28%) patients had a superficial wound complication, and it progressed to deep in 3 (3%) patients. Osteomyelitis was recorded in 1 (1%) patient, sternal dehiscence or dislocation in 20 (17%), rib resection in 1 (1%), and broken or removed steel wire in 6 (5%). In contrast, in the anterolateral thoracotomy group, 21 (16%) patients had a wound complication, 14 (11%) superficial, and 3 (2%) deep. Five (4%) underwent a reoperation for rib removal or fixation, and 1 (1%) patient had osteomyelitis. One (1%) patient in the anterolateral thoracotomy group suffered from a sternal fracture and dislocation after discharge from the hospital.

Sensitivity Analysis

The differences in primary outcome and transfusion data between groups can be found separately in the supplementary material for patients who did or did not need ECC during the LuTx, in Tables S17–S18 and S19–S20, respectively. In short, after multivariable analysis, for patients needing ECC during LuTx, the hospital stay was still significantly shorter for the anterolateral thoracotomy group despite a longer operation duration. Blood loss and transfusion parameters were no longer significantly different between groups (Table S18). For the patients operated without ECC, no differences were found in the primary outcomes between groups, except for a small increase in operation duration (Table S20).

Discussion

This paper sought to investigate differences in operative and clinical outcomes of patients undergoing bilateral LuTx via clamshell versus bilateral anterolateral thoracotomy, whilst elaborately correcting for confounding factors. Patients with an anterolateral thoracotomy experienced significantly less peri‐operative complications, such as lower blood loss and transfusion need. In addition, patients in the anterolateral thoracotomy group experienced less wound‐related complications. Moreover, these patients had less intraoperative ECC use, shorter ICU and hospital stay, and could be weaned from mechanical ventilation significantly faster. The use of intraoperative ECC, which was much higher in the clamshell group, seemed to play an important role in these differences in outcomes. Even though operating and ischemic times were slightly longer in the anterolateral thoracotomy group, survival of these patients was similar to the clamshell cohort. These results support the favorable effect of the bilateral anterolateral thoracotomy approach for LuTx on enhanced patient recovery compared to clamshell incision. Taking several aspects of the anterolateral thoracotomy technique into consideration, such as peri‐operative anesthesiologic and surgical management and learning curve, this technique can be considered as the standard technique of choice.

Operating Times

The possibility of prolonged operating time has been one of the arguments that favors the clamshell approach over the anterolateral thoracotomy incision. This study demonstrated that ischemic times were significantly longer for the anterolateral thoracotomy group, with the medians about 20 min apart in the uncorrected analysis. However, the clinical relevance of such a small difference is debatable. Additionally, the implant time, defined as the time between incision and reperfusion of the donor lung, was similar in both groups for the first lung, but approximately 30 min longer for the second lung in the anterolateral thoracotomy approach. This difference is presumably due to careful management of the implanted lung before initiation of single lung ventilation and perfusion, to which more attention was paid in the anterolateral thoracotomy group. This is done in order to reduce the sudden stress on the implanted lung as well as on the right ventricle, to prevent graft failure and to prevent the need for ECC, which was indeed much lower in our anterolateral thoracotomy group. Lastly, total operation duration was similar in the unadjusted analysis and slightly longer (36 min longer for an average patient) in the anterolateral thoracotomy group after multivariable correction, indicating that the time taken for careful recirculation of the first donor lung is made up for by not needing ECC, faster and effective hemostasis, and closure of the incision. Moreover, the learning curve for performing a LuTx through a bilateral anterolateral thoracotomy, which is admittedly more technically challenging, was not excluded for this analysis. In prior studies, one study also found a significantly longer ischemic time, but there was no difference between groups in the other three. All found a shorter total operation duration in the anterolateral thoracotomy group, however, no multivariable regression analysis was performed [3, 5, 7, 11]. Considering the results in our study and the reports from the literature, the concerns about the increasing operation duration do not seem to hold as a strong argument for the reluctance to switch from clamshell to bilateral anterolateral thoracotomy.

Blood Loss and Transfusion

Preferably, both blood loss and transfusion of blood products are kept to a minimum during LuTx surgery, in order to preserve scarce resources and prevent undesirable effects on LuTx outcome, such as transfusion‐related acute lung injury (TRALI), primary graft dysfunction, or transfusion‐associated circulatory overload (TACO) [10, 12, 13, 14]. In this study, significantly less blood loss was observed in the anterolateral thoracotomy group, reducing the need for intraoperative transfusion of erythrocytes and platelets. This was also found in two prior studies, whilst another showed no significant differences [5, 6, 7]. There are several possible explanations for having lower blood loss and transfusion need in the anterolateral thoracotomy group. Firstly, the anterolateral thoracotomy incision causes less surgical the incision is smaller, the sternum is left intact, and the internal mammary arteries and veins are not ligated, creating less possible sites to bleed from. Secondly, ECC use was much lower in the anterolateral thoracotomy group, and as a result, less patients are heparinized, reducing blood loss and transfusion need. Therefore, stratification for intraoperative ECC use was performed, after which these differences diminished or disappeared. Yet, the sample size also decreased, reducing the power of the analysis.

Intraoperative ECC Use

There are several points to be discussed regarding intraoperative ECC use in this study. Firstly, intraoperative ECC was used significantly more often in the clamshell group, almost twice as much as in the anterolateral thoracotomy group. In the literature, this finding is still a matter of discussion, with two studies confirming these results [15, 16] and two studies reporting otherwise [3, 6]. It is known that ECC use is associated with a varying effect on several of the outcomes, mostly in favor of the group without circulatory support [5, 6, 10, 17, 18]. This raises the question whether the use of ECC is a confounder or mediator of the effect between the incision type and these outcomes. In case it would only be a confounder, being independently associated with both exposure (incision) and outcome, it should be accounted for in multivariable regression analysis. However, in this study, we chose not to do so and instead add stratification in a sensitivity analysis, as we think intraoperative ECC use can be a mediator of the effect between incision and certain outcomes, and in that case, should not be corrected for [19]. The sternum is left intact with the bilateral anterolateral thoracotomy, possibly reducing extensive manipulation of the heart and mediastinum, which are unprotected in clamshell incision. However, performing a full mediation analysis was outside the scope of this study. Our stratified analysis showed that indeed, the differences in outcome between the groups partly rely on the use of intraoperative ECC.

Another point of discussion regarding intraoperative ECC in this context is that, in the case of the anterolateral thoracotomy, peripheral cannulation of CPB or ECMO is commonly used. One of the concerns when using peripheral VA‐ECMO or CPB in a patient with preserved cardiac output is that the patient can develop differential hypoxemia, a rare phenomenon which is often erroneously referred to as Harlequin anterograde native cardiac output of poorly oxygenated blood can mix with the retrograde flow of well‐oxygenated blood from the ECC, usually in the region of the aortic arch, causing poorly oxygenated blood to flow in the direction of the upper(right) half of the body including the brain [20]. This phenomenon can be prevented by a combination of actions, such as prevention of perfusion shunting during explanation of the native lung by clamping the pulmonary artery as soon as possible, continuous ventilation of the contralateral lung, explanting the most diseased lung first, and facilitating optimal venous drainage. Continuous monitoring of cerebral oximetry (INVOS) and measuring saturation on both arms are obligatory. Furthermore, a peripheral cannulation strategy comes with additional risk of complications of the groin and lower limb [21, 22]. However, peripheral cannulation does allow for mobilization in patients who require post‐operative ECMO [23]. Ultimately, the optimal ECC strategy will differ for each case, but the consensus is that peripheral cannulation is the optimal approach for patients operated through the anterolateral thoracotomy, which does have its advantages [15, 23, 24].

Postoperative Outcomes

Thirty‐day survival was comparable between both groups, which corresponds with findings by prior studies [5, 7, 16]. The LOS at the ICU and in the hospital were both significantly shorter for the anterolateral thoracotomy group in this study, even after multivariable analysis. A shorter ICU stay was also reported in three other studies [5, 6, 7], but the difference in hospital stay was not significantly different in two of those [5, 7]. Furthermore, the duration of mechanical ventilation was significantly in favor of the anterolateral thoracotomy group in our study. Similar results were reported in the literature [5, 6, 7]. In addition, rethoracotomies occurred less frequently in the anterolateral thoracotomy group in the unadjusted/univariable analysis; however, this was no longer statistically significant after multivariable logistic regression. Other factors that seemed to play a role in having to undergo a rethoracotomy in the postoperative phase were high urgency status, indication for transplantation, and size reduction (Table S12).

One of the major advantages of using the less invasive anterolateral thoracotomy approach is the reduction in wound‐related complications. Whereas 46% of the patients in the clamshell group experienced some form of a wound‐related complication, this was the case in only 16% of the anterolateral thoracotomy patients. The reduction in wound‐related complications was also noticed by other groups [3, 7, 16].

Additionally, several studies reported a higher postoperative (forced) vital capacity and forced expiratory volume in 1 s (FEV1) in the anterolateral thoracotomy group compared to the clamshell group. This observation persisted throughout the first year after the surgery [3, 5, 8, 11]. We did not include spirometry data in the current study.

Strengths and Limitations

Our study is one of the largest cohort studies investigating the effect of incision on lung transplant outcomes, with elaborate correction for confounding. However, it is subject to certain limitations. Firstly, this is a retrospective cohort study, for which data needed to be recovered from the electronic medical records, which makes it sensitive to bias. Although limited, some data were missing, particularly in patients from the earliest years of the study period. Moreover, some parameters were not recorded reliably, such as pain scores, quality of life, anastomosis time, or the occurrence of primary graft dysfunction. Other outcomes were used as a proxy for some of these variables, such as implant time for anastomosis time, and time to extubation, which is known to correlate well with PGD [25]. Secondly, since this study made use of a mainly historical cohort as control, the effect of time and changes in protocol cannot completely be ruled out. The two cohorts were heterogeneous in some baseline characteristics, probably partly due to advancements and evolving practices during the study period. For example, indications for LuTx changed over time. One example is the development of effective medical therapy for CF by means of cystic fibrosis transmembrane conductance regulator (CFTR) therapy, resulting in a decrease of CF diseased patients requiring LuTx. Moreover, the recipient age was extended from 65 years to 71 years, and more comorbidities in recipients were accepted prior to transplantation over time. Because of that, LuTx recipients became more high‐risk over time, but experience with these patients also increased as the number of transplants performed each year more than doubled over the study period. In addition, the LAS was introduced in 2014, which also influenced indications [26]. All patients with a LAS ≥50 were considered as HU for our analysis, but there will probably be some differences between the HU group before and after LAS implementation [27]. However, for those reasons, an elaborate correction for possible confounders was performed in an attempt to limit bias. Thirdly, LuTx remains a relatively rare type of surgery, and this is a single‐center study, so the sample size is limited. Furthermore, no correction for multiple testing was applied, hence, some of the differences may be statistically significant due to chance. Lastly, selection bias is probably present in this study. Especially during the transition period from clamshell to anterolateral thoracotomy, surgeons may have chosen to perform the anterolateral thoracotomy in the low‐risk recipients, whilst continuing with the more familiar clamshell incision for the high‐risk patients. This could also partly explain the difference in ECC use between groups. However, we have attempted to correct elaborately for this in the multivariable analysis.

Conclusion

In conclusion, performing bilateral sequential LuTx through the less invasive bilateral anterolateral thoracotomy without sternal division is a viable alternative compared to the traditional clamshell incision, despite technical difficulty. It is associated with only a limited increase in operating and ischemic times. While associated with similar survival, we report significantly less peri‐ and post‐operative complications and a faster recovery. Therefore, bilateral anterolateral thoracotomy can be considered as a standard technique for bilateral LuTx.

Author Contributions

Conceptualization: S.L., L.S., R.H., and E.M. Data S.L., M.H., and E.M. Methodology: S.L. and E.M. Formal S.L. Writing – original draft S.L. and E.M. Writing – review and S.L., L.S., R.H., J.B., M.H., and E.M. Visualization: S.L. and E.M. Supervision: E.M. All authors have read and agreed to the published version of the manuscript.

Funding

The authors have nothing to report.

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information