Background

Chronic obstructive pulmonary disease (COPD) is the most common indication for lung transplantation worldwide 1. Currently, lung transplantation is the final treatment strategy for patients with end-stage COPD. The points in favor of single lung transplantation (SLT) and double lung transplantation (DLT) are equivocal. However, researchers of some case–control series have reported better outcomes in patients who underwent DLT than in those who underwent SLT [1]; in their experience, SLT leads to a high rate of primary graft dysfunction. Conversely, a large retrospective registry analysis revealed equal outcomes between SLT and DLT [2]. We reviewed relevant published literature and noted two different opinions regarding SLT and BLT: most studies have indicated that DLT is better for survival than SLT, but others have provided data indicating equal outcomes between the two.

Therefore, we reviewed the existing literature on the subject and performed a meta-analysis of all included studies to determine whether SLT or DLT yielded better survival outcomes.

Methods

Search strategy and inclusion criteria

We searched the PubMed, Medline, and Scopus databases using one or more of the following keywords: “chronic obstructive pulmonary disease” and “single lung transplantation or double lung transplantation.” A total of 416 results were identified in the search. We excluded articles on animal studies; articles written in a language other than English; articles that were case reports, reviews, letters, and editorial comments; articles published before 2000; and articles on studies with less than 50 patients.

The primary inclusion criteria were that the study must compare two treatment arms, (i.e., SLT and DLT) and that all the included patients should have undergone lung transplantation for end-stage lung disease.

Data extraction and quality assessment

Two reviewers read all the included literature critically and extracted the relevant data, including the first author, year of publication, number of treatment arms, and survival results. The quality of the included studies was assessed by all authors using the Newcastle–Ottawa Scale, which comprises three parts for a case–control study or cohort study: “SELECTION” (four items), “COMPARABILITY” (one item), and “EXPOSURE” (three items). Disagreements between the two reviewers were resolved through discussions with the other authors, including the corresponding author.

Data synthesis and analysis

Patient survival was the primary outcome in this study. We used rate ratios to compare SLT and DLT. Some of the included studies used multiple variance analyses and presented data with hazard ratios; we also used these to compare SLT and DLT. A random effects model was used to pool individual rate ratios and hazard ratios. Heterogeneity was determined using I2 tests; I2 values of > 50% were considered indicative of obvious heterogeneity. Potential publication bias was determined using the Egger’s test and Funnel plots. Statistical significance was defined as P < 0.05. All statistical analyses were performed using the Comprehensive Meta-Analysis software, version 3 (Biostat, Englewood, NJ, USA).

Results

Study search and characteristics of the included patients

Overall, 416 records were identified through database searching. Two reviewers read the titles, abstracts, and keywords of these records, and selected 32 studies based on the inclusion and exclusion criteria (Fig. 1). These mostly comprised case–control studies and database analyses.

Fig. 1
figure 1

Preferred reporting items for systematic reviews and meta-analyses (PRISMA) flow diagram

Full size image

Seventeen of these were further excluded for the following reasons: patients with COPD were not separated from all lung transplant recipients [3,4,5,6,7,8,9,10]; survival data were not analyzed [11,12,13]; problems were noted with the statistical analyses, i.e., hazard ratios did not fit the 95% confidence intervals [14, 15]; SLT and DLT were not compared [16, 17]; a simulated model was used to compare the effects of SLT and DLT for COPD on waitlist outcomes, but long-term survival data were not reported [18]; and the OPTN/UNOS database was analyzed, but detailed survival data were not reported [19].

Finally, 15 studies remained for data analysis; most comprised retrospective case–control studies [1, 2, 20,21,22,23,24,25,26,27,28,29,30,31,32]. Some of these were single-center, retrospective case–control studies [1, 21, 23, 24, 28, 29]; the others were database cohort analyses [2, 20, 22, 25,26,27, 30,31,32]. The data extracted from all the included studies are provided in Table 1. We used the Newcastle–Ottawa Scale to appraise all the studies; the results are provided in Table 2.

Table 1 Patient characteristics, study methodology, and quality assessment of included trial
Full size table
Table 2 The detail of quality assessment of the included studies
Full size table

Pooled rate ratio and hazard ratio of survival

We analyzed the survival rate and compared the same between the SLT and DLT groups in each study. We also included the 1-year, 3-year, and 5-year survival data in the analysis. In some studies, results were obtained using multiple variance analyses and hazard ratios; we performed a separate analysis for these studies [2, 25, 27, 30].

The pooled rate ratios were 0.98 (P = 0.646; Fig. 2), 0.937 (P = 0.041; Fig. 3), and 0.775 (P = 0.000; Fig. 4) for the 1-year, 3-year, and 5-year survival, respectively.

Fig. 2
figure 2

Rate ratio analysis of 1-year survival following double and single lung transplantation. CI, confidence interval; DLT, double lung transplantation; SLT, single lung transplantation

Full size image
Fig. 3
figure 3

Rate ratio analysis of 3-year survival following double and single lung transplantation. CI, confidence interval; DLT, double lung transplantation; SLT, single lung transplantation

Full size image
Fig. 4
figure 4

Rate ratio analysis of 5-year survival following double and single lung transplantation. CI, confidence interval; DLT, double lung transplantation; SLT, single lung transplantation

Full size image

The pooled hazard ratio of survival was 0.857 (P = 0.388; Fig. 5a). Thabut et al. analyzed the International Society for Heart and Lung Transplantation database and reported different data after propensity score matching [2]. We included their study, with two different results, in the analysis because the P values were not significant. The pooled hazard ratio was 0.956 (P = 0.755; Fig. 5b).

Fig. 5
figure 5

Hazard ratio analysis of double and single lung transplantation. a Without propensity score matching in the study by Thabut et al. b With propensity score matching. CI, confidence interval; DLT, double lung transplantation; SLT, single lung transplantation

Full size image

The Egger’s test did not reveal a significant publication bias in the following: 1) pooled rate ratio analyses of the 1-year (P = 0.154), 3-year (P = 0.097), and 5-year (P = 0.242) survival; 2) hazard ratio analysis (P = 0.711); and 3) hazard ratio analysis with Thabut et al.’s propensity score matching results (P = 0.188). The Funnel plots are presented in Figs. 6, 7, 8, 9a and b.

Fig. 6
figure 6

Funnel plot of all studies that included 1-year survival data

Full size image
Fig. 7
figure 7

Funnel plot of all studies that included 3-year survival data

Full size image
Fig. 8
figure 8

Funnel plot of all studies that included 5-year survival data

Full size image
Fig. 9
figure 9

Funnel plot of all studies included in the hazard ratio analysis. a Without propensity score matching in the study by Thabut et al. b With propensity score matching

Full size image

Discussion

Patients with end-stage COPD are often recommended to undergo lung transplantation, which is currently the most acceptable treatment method. However, the debate between SLT and DLT still exists [33]. Lung transplantation is a complex and difficult surgery, and procedure selection is known to affect patient survival [34].

Our analysis showed that the early survival outcomes were equal between SLT and DLT. However, DLT achieved a better mid-term and long-term survival than SLT. The pooled hazard ratio did not reveal a significant difference between the two.

Most of the case–control series revealed a better outcome for DLT [1, 21, 24, 28, 29]. However, analysis studies based on a large registry revealed equal outcomes for both methods [2, 30]. Besides, DLT could bring about an organ shortage and increase the risk of mortality in patients on the waiting list. This is the primary reason the current review did not recommend whether patients with end-stage COPD should receive SLT or DLT.

The retrospective database study by Thabut et al. is an important one; it majorly contributed to the present meta-analysis due to its large sample size. Thabut et al. used different statistical methods (including propensity score matching) in an attempt to reduce the effect of confounding factors. They achieved the same result with these methods. We chose to include their study because we thought that their data, obtained with multiple methods, would allow us to better compare SLT and DLT.

The choice between DLT and SLT remains debatable. Waiting list mortality is major concern during choosing the appropriate procedure. SLT can reduce the waiting times associated with organ shortage [29]. However, DLT has been proven to yield better survival and quality of life outcomes in some studies [1]. This conflict will affect the choice of procedure, especially when the patient’s age is taken into consideration. DLT could provide a better quality of life for larger lung volumes [1]. For younger recipients, this is an important factor to consider while discussing the treatment plans with the transplantation team.

Our study had several limitations. First, all the included studies were case–control studies or retrospective analyses of registry data. Thus, the evidence level was not high. Several additional factors affect patient survival, including the patient’s age, center where the surgery is conducted and the facilities available there, and the surgeon’s experience and expertise. Two of the included studies involved age-based analyses [2, 20]; however, the meta-analysis pooled their data and masked the effect of age.

Furthermore, we excluded studies published before 2000 because surgery techniques and critical care have undergone significant changes in the past 20 years. The aforementioned factors would have affected our results had we included studies published before 2000 in our meta-analysis. Accordingly, we further excluded case–control studies with less than 50 patients since such low-volume studies could also affect our results.

It is impossible to conduct a prospective randomized trial on this subject due to ethical considerations regarding patient treatment. However, a retrospective registry analysis across multiple countries and comparison of the obtained results may provide data beneficial for patients with end-stage COPD worldwide. The more retrospective studies published, the more data we can collect for a meta-analysis to determine the different factors related to the outcomes of the two transplantation procedures.

Conclusions

We determined that in patients with end-stage COPD, DLT results in a better 3-year and 5-year survival than SLT.