INTRODUCTION

Size matching is an important factor that directly affects the outcomes of lung transplantation. An oversized lung allograft (OLA) is defined as a donor/recipient predicted total lung capacity (pTLC) ratio > 1.0^[1,2]. In practice, however, OLA identification can be used for any lung that does not fit into the thoracic cavity, with or without donor/recipient pTLC ratio > 1. The lungs are a highly susceptible organ and tend to be edematous, inflamed, and less compatible due to the effects of donor brain death, handling during procurement, cold storage, and ischemia-reperfusion injury^[3]. Additionally, a dilated right heart and/or an elevated diaphragm and/or an excessive intrathoracic fat tissue may occupy a space in thoracic cavity^[4]. Therefore, the lung allograft may become oversized due to one or more of these factors.

Primary closure of the chest after OLA use may cause undesirable complications such as atelectasis in the lungs, hemodynamic instability, primary graft dysfunction (PGD), venous occlusion, pneumonia, and anastomotic healing^[3-6].

Delayed chest closure (DCC) is one of the surgical maneuvering options that are administered to manage the size mismatch due to OLA. In plain definition, DCC is the closure of the open thorax without approaching the ribs and sternum. The primary goal here is to allow the lungs that do not fit into the thoracic cavity for various reasons to return to their original size and to prevent any compressive complications.

DCC is a well-defined method used in complicated cases of cardiac surgery since the 1970s^[7]. However, it does not have a long history and extensive knowledge in lung transplantation^[6,8]. Only a few serious case series have been published in the last decade^[4,8,9]. The aim of our study was to evaluate DCC results after lung transplantation in our clinic.

METHODS

Patients

Sixty patients who underwent lung transplantation between December 2016 and June 2019 were evaluated retrospectively. Patients who underwent bilateral lung transplantation and DCC for OLA were included in the study. Six patients who underwent single lung transplantation, four patients who underwent lobar transplantation, two patients who underwent retransplantation, and four patients who underwent DCC for hemodynamic instability were excluded from the study (Figure 1). Forty-four patients were divided into two groups as primary chest closure (PCC) (n=28) and DCC (n=16). Demographic characteristics, donor and operative characteristics, and results of the patients were compared.

Fig. 1 - Patient selection.

Download

Fig. 1 - Patient selection.

RESULTS

Patients’ Demographics

A total of 44 patients were included in the study. The mean age was 44.5 years and the male to female ratio was 25/19. The mean body mass index is 24.3 kg/m². PCC was performed in 28 patients and DCC was performed in 16 patients. The mean DCC duration was three days (range, 2 to 4). There was no significant difference between the groups in terms of demographic characteristics (P>0.05). Although the mean oxygen use, six-minute walk test, and mean pulmonary artery pressure (or PAPm) were higher in DCC group, there was no significant difference between the groups (P>0.05). The most common indication in both groups was idiopathic pulmonary fibrosis/interstitial lung disease. The comorbidities were similar in both groups (Table 1).

Table 1 - Patients’ demographics.

Parameter		Total (n=44)	PCC (n=28)	DCC (n=16)	P-value (<0.05)
	Age (years)	36.5±13.2	35.2±12.4	38.6±13.1	0.353
	Male	25	16	10	0.570
	Female	19	13	6	0.907
	BMI (kg/m2)	24.3±4.5	24.9±4.4	23.4±4.7	0.292
Respiratory state	Oxygen use at rest (lt)	4.2±2.0	3.8±1.6	4.9±2.4	0.077
	6MWT(m)	161±109	183±106	148±115	0.334
	FEV1 (%)	29.3±14.2	30.6±16.3	28±10.2	0.874
	FVC (%)	38.5±15.1	36.5±13.7	41.9±18.6	0.213
Cardiac state	Mean PAP (mm-Hg)	28.8±14.4	27.0±12.3	32.3±18.5	0.341
	TAPSE (mm)	2.0±0.6	2.2±0.5	2.0±0.8	0.235
	PVR (mm-Hg)	2.8±1.4	2.8±1.3	2.9±0.8	0.723
	Cardiac index (L/m/m2)	2.6±1.0	2.7±0.9	2.5±1.0	0.359
Indications	IPF	10	6	4	0.788
	ILD	13	8	5	0.853
	COPD	8	6	2	0.465
	Bronchiectasis	7	5	2	0.644
	Cystic fibrosis	5	3	2	0.859
	Sarcoidosis	1	0	1	0.186
Comorbidity	Diabetes mellitus	4	2	2	0.557
	Systemic arterial hypertension	6	4	2	0.870
	Previous thoracic surgery	9	6	3	0.834
	ECMO bridge to LTx	4	2	2	0.557
	Pleural empyema	1	0	1	0.186

6MWT=six-minute walk test; BMI=body mass index; COPD=chronic obstructive pulmonary disease; DCC=delayed chest closure; ECMO=extracorporeal membrane oxygenation; FEV1=forced expiratory volume in 1 second; FVC=forced vital capacity; ILD=interstitial lung disease; IPF=idiopathic pulmonary fibrosis; LTx=lung transplantation; PAP=pulmonary artery pressure; PCC=primary chest closure; PVR=pulmonary vascular resistance; TAPSE=tricuspid annular plane systolic excursion

Table 1 - Patients’ demographics.

Postoperative Course, Complications, and Outcomes

In the DCC group, extubation time (4.3 vs. 3.1 days, P=0.002) and ICU length of stay (7.6 vs. 5.2 days, P=0.016) were significantly higher than in the PCC group. The incidence of major complications was similar in both groups and there was no significant difference between the total number of major complications (P>0.05). Postoperative wound infection was significantly higher in the DCC group compared to the PCC group (18.6% vs. 0%, P=0.019).

The 90-day mortality was 18% in the DCC group and 17% in the PCC group (P>0.05). Acute rejection episodes were observed in 31% of the DCC group and 39% of the PCC group at one-year follow-up (P>0.05). Median survival was 14 months in all patients. There was no significant difference in survival between the PCC and DCC groups (16 vs. 13 months, P=0.300) (Table 2).

Table 2 - Donor characteristics, operative data, and outcomes.

Parameter		Total (n=44)	PCC (n=28)	DCC (n=16)	P-value
Donor age		35±12	33±12	38±14	0.175
Donor gender	Male donor	23	13	10	0.310
	Female donor	21	15	6	0.310
	Donor PaO2 (mm-Hg)	397±87	404±83	385±96	0.442
	Donor/recipient pTLC rate	0.99±0.1	0.96±0.1	1.06±0.8	0.008*
Allograft ischemia time	Ischemia of first lung (minute)	255±47	262±55	242±25	0.261
	Ischemia of second lung (minute)	420±56	422±63	416±44	0.582
	Total operation time (minute)	542±51	549±60	530±27	0.322
	Total blood product use	7.1±3.4	6.7±2.9	7.7±4.1	0.113
	Intraoperative ECMO use	17	9	8	0.822
Additional procedure	Wedge resection	6	4	2	0.870
	Lobectomy	4	4	0	0.117
	Diaphragm plication	5	2	3	0.249
	Excessive fat tissue excision	2	0	1	0.186
	Extubation time (day)	3.5±3.2	3.1±3.3	4.3±2.9	0.002*
	ICU time (day)	6.1±4.1	5.2±3.0	7.6±5.3	0.016*
Complications	PGD requiring ECMO	6	4	2	0.644
	Bleeding requiring re-exploration	2	2	0	0.279
	Tracheostomy	5	4	1	0.424
	Arrhythmia requiring treatment	6	4	2	0.870
	Bronchopleural fistula	1	1	0	0.450
	Renal insufficiency requiring dialysis	3	3	0	0.624
	Cerebrovascular event	1	1	0	0.450
	Pleural empyema	3	2	1	0.264
	Total major complication	27	21	6	0.312
	Wound infection	3	0	3	0.019*
	Acute rejection episode	16	11	5	0.598
	90-day mortality	8	5	3	0.942
	Median survival (mo)	14	16	13	0.300

DCC=delayed chest closure; ECMO=extracorporeal membrane oxygenation; ICU=intensive care unit; PaO2=partial oxygen pressure; PCC=primary chest closure; PGD=primary graft dysfunction; pTLC=predicted total lung capacity

* Significant P-value.

Table 2 - Donor characteristics, operative data, and outcomes.

DISCUSSION

Our results showed that DCC maneuver for bilateral lung transplantation is as safe and effective as PCC in terms of results. Although wound infection in DCC is more common than in PCC, and ICU length of stay and extubation time are longer in DCC than in PCC, early major complications, mortality, one-year acute rejection attacks, and survival were comparable with PCC.

Similar results have been reported in other important case series. Force et al.^[7] reported the first case series on DCC in 2006. In 28 lung transplantations, seven (25%) patients underwent DCC and were compared with 21 patients who underwent PCC. The mean DCC time was 5.3 days (range, 3 to 7). The DCC group showed significantly higher blood transfusion requirements, higher pulmonary artery pressure, greater cardiopulmonary bypass (CPB) usage, longer ischemia time, and a lower partial oxygen pressure/fraction of inspired oxygen, or PaO₂/FiO₂, ratio than the PCC group. PGD development and tracheostomy rate were higher and length of hospital stay was longer in the DCC group, but no difference was observed in one-month mortality. The authors argued that the method may be an important alternative strategy in patients with PGD findings. D’Cunha et al.^[13], between 2006 and 2008, performed DCC in five patients. One patient died due to bowel obstruction at five months. The average discharge time was 41 days (range, 26 to 62). The author reported that ECMO and its associated costs and serious complications can be avoided in unstable patients with DCC. In the largest case series of DCC published by Shigemura et al.^[4], 90 (10.3%) of 873 patients who underwent lung transplantation between 2004 and 2017 were performed DCC and their results were compared with the PCC group. The mean DCC duration is 4.5 days (range, 1 to 18). In DCC, operative time, CPB use, and duration and transfusion were higher. They reported that PGD requiring ECMO support, one-year acute rejection episode, and 30- and 90-day mortality were higher in the DCC group. In a study by Rafiriou et al.^[8], between 2009 and 2016, 51 (7%) of 770 patients who underwent lung transplantation were performed DCC and were compared with PCC. The average DCC duration was 4.6±2.3 days. The prolonged intubation was higher in the DCC group, but the results were similar in terms of other complications and long-term survival. Case series in the literature were given in Table 3.

Table 3 - Case series in the literature.

In addition to OLA, DCC may also be required in cases of acute pulmonary edema, hemodynamic instability, and high risk of bleeding^[4,9]. The indication for DCC depends on the choice of the surgeon intraoperatively. In OLA, there is a size mismatch, which needs to be handled intraoperatively. In this PCC case, the surgeon has two options: allograft volume reduction (lung resections) and/or expansion of the thoracic volume (excision of excessive intrathoracic fat tissue and diaphragmatic plication). However, surgical resection of lung volume should always be the last resort. Instead, DCC maneuver may be certainly a more reasonable solution to allow the lung to shrink spontaneously. Thus, the allograft can reach its normal size and the problem can be solved with a smaller process. In our clinic, we routinely perform DCC first in the management of OLA. After two to four days of waiting period, if the lungs fit into the thoracic cavity, we do PCC. If the lungs still do not fit, we first perform maneuvers (diaphragm plication and/or excision of excessive intrathoracic fat tissue) to enlarge the thoracic cavity. If the lungs still do not fit, resection is the last resort. We did not need lobectomy in any of the patients who underwent DCC with this application.

Technically, it is possible to perform three methods for DCC after lung transplantation. Type 1, simple skin approximation using continuous suture^[4]; Type 2, the chest is left open and a layer of latex-free Esmark bandaging (Fulflex Elastomerics Worldwide, Lincoln, Rhode Island, United States of America) is attached to the skin using sutures^[4,7,13]; and Type 3, the sternum is retracted by a rib spreader and then the second method is performed^[4,7]. We always used simple skin approximation in all patients. We never needed the other two types of method. Shigemura et al.^[4] reported that Types 2 and 3 DCC techniques were associated with a higher risk of death, whereas in the Type 1 DCC technique, the risk of PGD requiring ECMO, renal insufficiency, and death was reduced compared to PCC. In multivariate analysis, prolonged CPB duration, postoperative ECMO requirement, and Type 3 DCC technique increased the risk of death in DCC. The findings of Shigemura partially explain why our results are slightly better than of some other reports.

Whether DCC increases the risk of infection is controversial. Aguilar et al.^[9] performed DCC in 67 (29%) of 232 patients who underwent lung transplantation between 2010 and 2014. In 22 (9%) of the transplanted patients, infection developed at the surgical site. Eighteen of them were wound infection, eight were pleural infection, and four were concomitant wound and pleural infection. Patients with DCC had significantly more infections than those with PCC (19% vs. 5%, P=0.001). In multivariate analyses, DCC was found to be an independent risk factor for surgical site infections. However, Force et al.^[7], Shigemura et al.^[4], and Rafiriou et al.^[8] reported in their published case series that DCC did not increase the risk of infection. In our case series, surgical site infection (three wound infections and three pleural empyemas) was detected in six (10%) patients. Supporting the results of Aguilar, only wound infection was significantly higher in patients undergoing DCC compared to PCC patients (18% vs. 0%, P=0.019).

The use of OLA has been reported to reduce the risk of PGD and increase long-term survival after lung transplantation^[1,2]. However, Shigemura et al.^[4] reported that after a lung transplant using an OLA, the primary PCC disrupted the allograft hemodynamics and physiology, exacerbating existing lung injury and increasing the risk of developing PGD. They reported that the development of PGD requiring ECMO support was significantly higher in patients undergoing DCC than in those undergoing PCC (31% vs. 2.8%, P<0.05). However, the development of PGD varied with the technique of DCC. PGD development rate after DCC with simple skin closure technique (Type 1) was 9.6%, and it was 60% and 62% for the second and third techniques (Types 2 and 3), respectively. The authors concluded that PGD can be reduced by appropriate technique and careful post-DCC management. However, they did not explain why DCC with the Type 1 technique leads to less PGD. Force et al.^[7] detected moderate or severe PGD findings in five of the six patients who underwent DCC and the frequency was significantly higher than that of PCC (P=0.0002). Aguilar et al.^[9] found that postoperative grades 2 and 3 PGD development was higher in patients with DCC compared to those with PCC. In our case series, total PGD was 10% and there was no significant difference between the groups. The fact that our DCC case series covers only OLA and that we used only simple skin closure technique (Type 1) in all patients may have affected our results.

The timing and criteria of final closure of the chest of patients undergoing DCC are not fully defined. The average DCC duration is 4.5 to 6 days (range, 1 to 18) in publications^[4,7-9,13]. In DCC lasting more than three days, exploration and washout are performed in every three days in the operating room and this is continued until final closure. Aguilar et al.^[9] reported median DCC duration as two days and median washout number as one. In our practice, the average DCC time was three days (range, 2 to 4) and no patient needed recurrent explorations. At the end of the DCC period, the chest was reopened, and exploration and washout were performed. The criteria for final closure after DCC are not well defined. D’Cunha et al.^[13] think that it is sufficient to have a central venous pressure > 10 mmHg and an acceptable low hemodynamic support for final closure. Shigemura et al.^[4] recommend that the patients have good renal function and diuresis, the allograft is dry, the coagulation parameters are normal, and the DCC duration is > 72 hours. Also, they recommend being patient and do not rush for final closure. Although there are no definite indications for final closure after DCC in our clinic, it is generally sufficient for us to have dry graft on chest radiography and to be hemodynamically stable. However, we agree with Shigemura's recommendation that the final closing should be done at least 72 hours later.

Limitations

This is a single-center retrospective cohort study and the data were taken from the patients’ files. Therefore, the data of some patients may be overlooked or not objective enough. Since the sample size was small, a general statistical comparison was made, and detailed statistical models could not be studied. However, only patients undergoing DCC for OLA were included in the study and the cohort was attempted to be homogenized.

CONCLUSİON

We believe that DCC contributes to the solution of the problem of mismatch caused by OLA in lung transplants without increasing the major complications and mortality. However, further studies are needed to compare DCC with other applications in the management of mismatching caused by OLA.