The impact of segmentectomy versus lobectomy on pulmonary function in patients with non-small-cell lung cancer: a meta-analysis

Objective Segmentectomy has been reported as an alternative to lobectomy for small-sized NSCLC without detriment to survival. The long-term benefits of segmentectomy over lobectomy on pulmonary function have not been firmly established. This meta-analysis aims to compare postoperative changes in pulmonary function in NSCLC patients undergoing segmentectomy or lobectomy. Methods Medline, Embase, Web of Science and Scopus were searched through March 2021. Statistical comparisons were made when appropriate. Results Fourteen studies (2412 participants) out of 324 citations were included in this study. All selected studies were high quality, as indicated by the Newcastle–Ottawa scale for assessing the risk of bias. Clinical outcomes were compared between segmentectomy and lobectomy. ΔFEV1 [10 studies, P < 0.01, WMD = 0.40 (0.29, 0.51)], ΔFVC [4 studies, P < 0.01, WMD = 0.16 (0.07, 0.24)], ΔFVC% [4 studies, P < 0.01, WMD = 4.05 (2.32, 5.79)], ΔFEV1/FVC [2 studies, P < 0.01, WMD = 1.99 (0.90, 3.08)], and ΔDLCO [3 studies, P < 0.01, WMD = 1.30 (0.69, 1.90)] were significantly lower in the segmentectomy group than in the lobectomy group. Subgroup analysis showed that in stage IA patients, the ΔFEV1% [3 studies, P < 0.01, WMD = 0.26 (0.07, 0.46)] was significantly lower in the segmentectomy group. The ΔDLCO% and ΔMVV% were incomparable. Conclusion Segmentectomy preserves more lung function than lobectomy. There were significantly smaller decreases in FEV1, FVC, FVC%, FEV1/FVC and DLCO in the segmentectomy group than in the lobectomy group. Supplementary Information The online version contains supplementary material available at 10.1186/s13019-022-01853-3.


Introduction
Lung cancer is one of the leading cause of cancer-related death worldwide [1]. Surgical resection for non-small-cell lung cancer (NSCLC) is the standard treatment that leads to the best chance of a cure. For the 100-year history of surgery, lobectomy has remained the gold standard for operable NSCLC. In recent years, segmentectomy has been reported as an alternative to lobectomy for smallsized NSCLC without detriment in survival [2]. Theoretically, segmentectomy has an advantage over lobectomy on anatomical functional. However, the long-term benefits of segmentectomy over lobectomy on pulmonary function have not been firmly established. Reports related to the utility of segmentectomy in preserving lung function are conflicting. The purpose of this study was to perform a meta-analysis to compare postoperative changes in pulmonary function in NSCLC patients undergoing segmentectomy or lobectomy.

Inclusion and exclusion criteria
Studies were included if they met the following inclusion criteria: (1) patients diagnosed with NSCLC underwent surgical treatment; (2) comparative data between segmentectomy and lobectomy were available; and (3) preoperative and postoperative pulmonary function data were available. The outcomes included forced expiratory volume in 1 s (FEV1), predicted FEV1 percentage (FEV1%), forced vital capacity (FVC), predicted FVC percentage (FVC%), FEV1/FVC, maximal voluntary ventilation (MVV), diffusion capacity of carbon monoxide (DLCO) and predicted DLCO percentage (DLCO%). Studies were excluded if the full text was not in English or could not be accessed.

Search strategy
Medline, Embase, Web of Science and Scopus were searched through May 2021. The following search terms and strategies were used: (1) respiratory function OR pulmonary function OR FEV1 OR FVC OR MVV OR DLCO; (2) lung cancer; (3) lobectomy AND (segmentectomy OR sublobar resection OR limited resection), and (1) AND (2) AND (3). Data were extracted with a standardized form. The Newcastle-Ottawa Scale (NOS) was used for quality assessment.

Statistical analysis
Inconsistency between studies was quantified by calculating the I 2 statistic. Continuous variables were reported as weighted mean differences (WMDs) and 95% confidence intervals (95% CIs). A random-effects model was used for heterogeneous data (I 2 > 50%), whereas a fixedeffects model was used for homogenous data (I 2 < 50%). P < 0.05 was considered to be statistically significant. SAS software, version 9.1 (SAS Institute, Cary, NC, USA) and Review Manager, version 5.4 (The Cochrane Collaboration) were used to perform statistical analysis.

Search results
The initial database search identified 172 articles in Medline, 251 in Embase, 154 in Web of Science and 166 in Scopus. After excluding duplicate records, 324 studies were included. A total of 273 articles were excluded because they failed to meet the inclusion criteria after review of the abstracts and titles. An additional 37 articles were excluded after the full text review. Hence, a total of 14 studies (13 retrospective and 1 prospective observational) including 2412 patients (976 sublobar resection and 1436 lobectomy) were finally selected. The detailed selection process is shown in Fig. 1.
The major characteristics of the participants in the included studies are shown in Table 1. Patients in 8 studies had NSCLC with TNM stage I. Six studies included the VATS approach only, 2 included the open approach only, and 3 included both. Two studies included wedge resections. The follow-up time (from surgery to postoperative pulmonary function) ranged from 3 to 60 months.

Quality assessment
The quality of the included studies was assessed using the NOS (www. ohri. ca/ progr ams/ clini cal_ epide miolo gy/ oxford. htm). Two independent reviewers conducted the assessment. Disagreements were resolved by discussion. Of the studies, seven scored 9 points, four scored 8 points, two scored 7 points, and one scored 6 points, indicating that all the studies had relatively high quality ( Table 2).

Discussion
Previous studies suggested that segmentectomy confers little functional advantage over lobectomy [10]. It was concluded that lobectomy should remain the procedure of choice despite the slight functional advantage of limited resection. In the present study, we compared postoperative changes in pulmonary function in patients undergoing segmentectomy or lobectomy. This metaanalysis showed that there were significantly fewer decreases in FEV1, FVC, FVC%, FEV1/FVC and DLCO in the segmentectomy group than in the lobectomy group. Subgroup analysis also showed that the decrease in FEV1% was significantly less in the segmentectomy group in stage IA patients. Altogether, these studies support the assumption that segmentectomy preserves more lung function than lobectomy.  Pulmonary function tests are recommended in all patients who undergo thoracic surgery [16]. Theoretically, segmentectomy has an anatomical functional advantage over lobectomy. First, as the adult lung cannot regenerate new alveolar septal tissues, postoperative pulmonary function is mainly determined by the amount of lung resected. Second, anatomical excursion of the nonoperated lobe after lobectomy occurred. For example, a right upper lobectomy will damage the function of the middle lobe due to the kink of the middle lobar bronchus and pulmonary artery [17]. Third, compensatory lung growth could already have occurred in the ipsilateral nonoperated lobe in the lobectomy group before the operation due to the decreased function in the operated    lobe, resulting in less space for postoperative lung growth [5].
FEV1 is an indicator of airway resistance. Changes in FEV1 are largely related to ventilation mechanisms, including existing airway obstruction, compensatory expansion of the residual lung, and chest wall activity [18]. Lung resection will inevitably lead to displacement of the remaining lobe. The meta-analysis showed that the decrease in FEV1 was higher in the lobectomy group, indicating that lobectomy is more likely to increase airway resistance. A recent meta-analysis conducted in early-stage NSCLC patients concluded that segmentectomy conserves more FEV1 (5 studies, 933 patients), %FEV1 (5 studies, 976 patients) than lobectomy (28). Changes in the FVC are mainly determined by the amount of lung tissue resected. After lung resection, the remaining part of the lung expands and compensates for the resected lobe [19]. The meta-analysis showed that both FVC and FVC% were more rapidly improved in the segmentectomy group, indicating that segmentectomy has an advantage in the preservation of lung volume. FEV1/FVC is an essential parameter to phenotype the functional pattern of patients if obstructive, restrictive or normal [20]. There were only 2 studies reporting changes in FEV1/FVC. The meta-analysis showed that the ΔFEV1/FVC was lower in the segmentectomy group. DLCO reflects the capillary surface area available for gas diffusion. Preoperative DLCO has been demonstrated to predict the risk of complications, short-and long-term outcomes and the length of hospitalization in patients undergoing thoracic surgery [21]. The meta-analysis showed a lower degree of DLCO decrease in the segmentectomy group, indicating that it had better preservation of oxygenation.
Pulmonary function after lung resection can be affected by a number of factors. The number of resected segments is an important factor. Several studies observed a positive relationship between the number of resected segments and the loss of pulmonary function [5,12,22]. As each lobe consisted of different numbers of segments, the improvement of pulmonary function was also determined by the resected lobe. Therefore, Macke et al. classified patients into the following two groups: those who had 1-2 segments resected and those who had 3-5 segments resected [12]. This classification could reduce the influence of different lobes and could be adopted in future studies. Furthermore, anatomical excursion of the remaining lobe could also influence the preservation rate of the residual lobe. As mentioned above, right upper lobectomy can cause a reduction in the volume of the right middle lobe [23]. Tane et al. found that residual lobe function was the most preserved after S6 segmentectomy, suggesting that the shape of the preserved segments (basal segment) may be amenable to inflation without anatomic displacement [7].
Emphysema could also affect postoperative pulmonary function. Kashiwabara et al. reported that there were some patients with emphysema receiving lobectomy who had a greater advantage in postoperative pulmonary functions than segmentectomy [11]. It was speculated that the removal of an emphysematous parenchyma may have caused a partial improvement of the regional lung volume distribution and ventilation inhomogeneity, thus causing 'compensatory lung growth' . However, the selected studies rarely described whether they included patients with emphysema or chronic obstructive pulmonary disease (COPD), which might result in increased heterogeneity.
The influence of the surgical approach on pulmonary function is controversial. Some researchers reported that no differences were found between VATS surgery and open surgery [24,25]. In contrast, some studies showed low functional loss after VATS segmentectomy, indicating that the functional benefit of segmentectomy may add to that of VATS [4,6,15]. The selected studies in this meta-analysis contained both VATS and open approaches. Subgroup analysis failed due to the high heterogeneity. More data are needed to achieve a convincing conclusion.
The influence of follow-up time on the recovery of pulmonary function was small. Koike et al. showed that postoperative VC and FEV1 gradually increased within 3 months of surgery and remained stable thereafter [26]. Similarly, Kobayashi et al. found that the VC% and FEV1% remained almost the same 1 year after surgery [9]. It was suggested that the decreases in VC and FEV1 are caused by ageing and are not affected by the operation [27].
Our study has several limitations. The lack of prospective studies influences the data quality. In addition, several factors (e.g., smoking status, complications, surgical procedure, pathological type, adjuvant therapy, and patient effort in pulmonary function tests) that may influence pulmonary function were not included in the selected studies, adding to the heterogeneity. Third, only English literature was included in our study. We also found several articles written in Japanese, Turkish or Chinese when searching for studies. This meta-analysis may be more broadly representative if we include studies in all languages.

Conclusion
This meta-analysis suggests that segmentectomy preserves more lung function than lobectomy. There were significantly smaller decreases in FEV1, FVC, FVC%, FEV1/FVC and DLCO in the segmentectomy group