The volume threshold of 300 versus 100 ml/day for chest tube removal after pulmonary lobectomy: a meta-analysis

The volume threshold of 300 versus 100 ml/day for chest tube removal after pulmonary lobectomy: a... Abstract OBJECTIVES In this meta-analysis, we conducted a pooled analysis of clinical studies comparing the efficacy of a volume threshold of 300 ml/day before removing a chest tube (CT) versus 100 ml/day after a lobectomy. METHODS According to the recommendations of the Cochrane Collaboration, we established a rigorous study protocol. We performed a systematic electronic search of PubMed, Embase, Cochrane Library, Web of Science databases, CNKI, the Wanfang database, CBMdisc and Google Scholar to identify articles to include in our meta-analysis. A literature search was performed using relevant keywords. A meta-analysis was performed using RevMan© software. RESULTS Five studies, published between 2014 and 2015, including 615 patients (314 patients who had the CT removed when daily drainage was <300 ml and 301 patients who had the CT removed when daily drainage was <100 ml) met the selection criteria. From the available data, the patients using the volume threshold of 300 ml/day had a significantly decreased duration of drainage [MD = −44.07; 95% confidence interval (CI) −64.45 to −23.68; P < 0.0001] and hospital stay after operation (MD = −2.25; 95% CI −3.52 to −0.97; P = 0.0006) compared with patients using a volume threshold of 100 ml/day after a pulmonary lobectomy. However, no significant differences were observed in postoperative complications, such as pleural fluid reaccumulation [Odds ratio (OR) = 1.73; 95% CI = 0.74–4.07; P = 0.21] and atelectasis (OR = 0.97; 95% CI = 0.52–1.81; P = 0.93). Thoracentesis rates after removing the CT also showed no significant difference (OR = 1.53; 95% CI 0.55–4.22; P = 0.41). CONCLUSIONS Our results showed that a higher volume threshold, up to 300 ml/day, is effective in reducing hospitalization times and duration of drainage in patients who undergo a lobectomy. Moreover, the volume threshold of 300 ml/day does not increase the occurrence of postoperative atelectasis, pleural fluid reaccumulation and thoracentesis rates. However, this review is limited by the methodological quality of the included trials, and additional studies according to the recommendations of Cochrane Library are appreciated. Chest tube, Lobectomy, Drainage, Volume threshold, Meta-analysis INTRODUCTION The need for chest tube (CT) drainage is common after operations on the pleural space or chest trauma. In general, the use of CT could help expend postoperative residual lungs and drain the fluid and air from the pleural space. In addition, CT could also be an indication of prolonged air leaks, haemothorax, empyema and chylothorax. CTs are known to have a long history of clinical use; however, indications for its removal are often controversial, including parameters such as volume threshold, drainage time and protein content of the pleural fluid [1]. There is wide variation in the mean duration of CT drainage among several institutions. In general, the standard for removing the tube requires no air leakage or haemothorax and fluid drainage of <100 ml for at least 24 h after a lobectomy [2]. Recently, several studies have reported the safe removal of the CT with drainage exceeding 100 ml for at least 24 h, and the new volume threshold of 300 ml/day has been used in many randomized trials [3–7]. Although some recent randomized trials have compared the effectiveness of using a volume threshold of 300 or 100 ml/day before removing the CTs, few available data could support which algorithm is more effective. There are no evidence-based consensus recommendations for the optimal volume threshold to be used. The main objective of this meta-analysis is to conduct a pooled analysis of clinical studies to compare the volume threshold of 300 ml/day with 100 ml/day before removing the CT with a statistical power much higher than that in each trial. Objective outcomes such as duration of drainage, volume of drainage, postoperative hospital stay, pleural fluid reaccumulation, thoracentesis rate and complications after pulmonary lobectomy were collected and analysed. PATIENTS AND METHODS A rigorous study protocol was established according to the recommendations of the Cochrane Collaboration. Inclusion criteria Types of studies All published RCTs comparing volume thresholds of 100 and 300 ml/day before removing a CT after pulmonary lobectomy were considered eligible for the meta-analysis. Language was limited to English and Chinese. Abstracts or unpublished data were included only if sufficient information on interventions and outcomes was available and if the final results were confirmed by contact with the first author. Types of participants The following RCTs were included: Trials comprising patients who were scheduled to undergo the pulmonary lobectomy or bilobectomy operation. Trials comparing volume thresholds of 100 and 300 ml/day before CT removal after pulmonary lobectomy. Trials in which the primary outcome measure was the duration of drainage or the amount of drainage. The secondary outcome measures were hospital stay after operation, postoperative complications or the number of patients who need thoracentesis after removing the CT. Exclusion criteria Patients who suffer from postoperative prolonged air leakage. Patients who suffer from densely bloody, purulent or cloudy pleural effusion after operations. Patients who were expected to have increased postoperative haemorrhage such as haematological system diseases and pleural extensive adhesion. Patients who suffer from adverse cardiovascular and cerebrovascular events. Literature search The Cochrane Central Register of Controlled Trials in the PubMed, Embase, Cochrane Library, Web of Science databases, CNKI, Wanfang database and CBMdisc were searched for RCTs comparing different volume thresholds before removing CTs after pulmonary lobectomy without language restriction. Moreover, Google Scholar, Baidu Scholar and reference lists of all the included studies were searched for additional reports. Contact with the authors was initiated by e-mail or telephone if any information was not available. The search strategies used the following major terms: ‘Pneumonectomy[Mesh]’, ‘pulmonary lobectomy[Mesh]’, ‘bilobectomy’, ‘Chest Tubes[Mesh]’, ‘Drainage[Mesh]’, ‘Randomized Controlled Trial OR random’, ‘Pneumonectomy OR Lung Volume Reduction OR pulmonary lobectomy OR Lobectomy OR bilobectomy OR pulmonary resection OR lung resection’ and ‘chest drain OR pleural drainage OR thoracic drainage OR Chest Tube’. Data extraction and quality assessment The described search strategy was used to obtain titles and abstracts of RCTs that were relevant to this review. Two reviewers independently assessed the titles and abstracts of all identified trials to confirm the fulfilment of inclusion criteria, and the duration of drainage, volume of drainage in the first 24 h, postoperative hospital stay, pleural fluid reaccumulation, thoracentesis rate and complications after pulmonary lobectomy were collected; data abstraction was performed independently by 2 reviewers. Any difference of opinion or disagreement that arose in the course of searching, data abstraction, quality assessment or other related work between the 2 investigators was resolved by discussion. Quality assessment was performed after full-text reading, and trials of low quality would be excluded. The risk of bias of the included RCTs was assessed according to Cochrane Handbook 5.1.0 and the 5 criteria used were listed in Table 1. Each entry was definitively judged by an answer (yes/no/unclear), which stands for a low, a high and an unclear or unknown risk of bias [8]. Table 1: Risk of bias in included studies Author Adequate sequence generation Allocation concealment Blinding Incomplete outcome data Free of selective reporting Free of other bias Han et al. [5] Yes (table of random numbers) Unclear Unclear Yes Unclear Unclear Zhang et al. [7] Yes (table of random numbers) Unclear Unclear Yes Unclear Unclear Zhang et al. [4] Yes (table of random numbers) Unclear Unclear Yes Unclear Unclear Li et al. [6] Unclear Unclear Unclear Yes Unclear Unclear Wang et al. [3] Yes (table of random numbers) Unclear Unclear Yes Unclear Unclear Author Adequate sequence generation Allocation concealment Blinding Incomplete outcome data Free of selective reporting Free of other bias Han et al. [5] Yes (table of random numbers) Unclear Unclear Yes Unclear Unclear Zhang et al. [7] Yes (table of random numbers) Unclear Unclear Yes Unclear Unclear Zhang et al. [4] Yes (table of random numbers) Unclear Unclear Yes Unclear Unclear Li et al. [6] Unclear Unclear Unclear Yes Unclear Unclear Wang et al. [3] Yes (table of random numbers) Unclear Unclear Yes Unclear Unclear Table 1: Risk of bias in included studies Author Adequate sequence generation Allocation concealment Blinding Incomplete outcome data Free of selective reporting Free of other bias Han et al. [5] Yes (table of random numbers) Unclear Unclear Yes Unclear Unclear Zhang et al. [7] Yes (table of random numbers) Unclear Unclear Yes Unclear Unclear Zhang et al. [4] Yes (table of random numbers) Unclear Unclear Yes Unclear Unclear Li et al. [6] Unclear Unclear Unclear Yes Unclear Unclear Wang et al. [3] Yes (table of random numbers) Unclear Unclear Yes Unclear Unclear Author Adequate sequence generation Allocation concealment Blinding Incomplete outcome data Free of selective reporting Free of other bias Han et al. [5] Yes (table of random numbers) Unclear Unclear Yes Unclear Unclear Zhang et al. [7] Yes (table of random numbers) Unclear Unclear Yes Unclear Unclear Zhang et al. [4] Yes (table of random numbers) Unclear Unclear Yes Unclear Unclear Li et al. [6] Unclear Unclear Unclear Yes Unclear Unclear Wang et al. [3] Yes (table of random numbers) Unclear Unclear Yes Unclear Unclear Statistical analysis Statistical analysis for the meta-analysis was performed with RevMan (Review Manager Version 5.3.3 for windows; Cochrane Collaboration, Oxford, UK, 2014). A pooled relative risk was calculated with 95% confidence intervals (CIs). Odds ratios (ORs) or risk ratios were used for dichotomous variables, and a mean difference was used for continuous variables. χ2 heterogeneity tests with statistical significance at P < 0.10 were applied to test statistical heterogeneity among trials. The I2 statistics were also applied to assess the proportion of variability in the results attributable to heterogeneity across studies, with I2 < 50% considered as low-level heterogeneity and I2 > 50% as high-level heterogeneity. Fixed-effects model was used if there was no statistically significant heterogeneity (P ≥ 0.10, I2 < 50%), and the random-effects model was used if there was significant heterogeneity (P < 0.10, I2 > 50%). Because the characteristics of the patients, operation methods, number of CTs used after an operation and other confounding factors were not consistent among studies, we further conducted sensitivity analyses to explore possible explanations for heterogeneity. Sensitivity analysis was carried out by omitting 1 trial in each turn. The sensitivity analysis did not materially alter the pooled results, adding robustness to our findings. Intention-to-treat analysis was not performed because of insufficient information about loss to follow-up in treatment and control groups. A funnel plot was not used because of the limited number of RCTs. RESULTS Study description The flowchart for literature screening is presented in Fig. 1. According to the established search strategy that was used, a total of 838 potentially relevant literature items were identified in the databases, and 27 duplicate studies were excluded by NoteExpress software. After screening the titles and abstracts of the remaining 791 studies, 659 irrelevant studies were excluded, leaving 132 studies for further assessment. After the full-text review of the 132 studies, we excluded 127 of them. Therefore, 5 trials that fulfilled the inclusion criteria were included [8]. Figure 1: View largeDownload slide Flow diagram: selection and evaluation process of the eligible studies. Figure 1: View largeDownload slide Flow diagram: selection and evaluation process of the eligible studies. Characteristics of the included studies The characteristics of the included studies are presented in Table 2; all 5 trials were performed in China. Both groups were well matched at baseline from the information on all trials. Table 2: General characteristics of included studies Author Study sites Year Number of patients Cases Male/female (n/n) Age (years) 100a 300b 100 300 100 300 Han et al. [5] China 2015 260 90 83 56/34 49/34 60.23 ± 9.61 61.22 ± 13.38 Zhang et al. [7] China 2014 70 29 41 14/15 21/20 55 ± 12 56 ± 14 Zhang et al. [4] China 2015 166 78 88 52/26 56/32 58.7 ± 3.2 58.4 ± 2.6 Li et al. [6] China 2014 98 50 48 Nullc Nullc Nullc Nullc Wang et al. [3] China 2015 108 54 54 35/19 37/17 70.45 ± 5.31 71.44 ± 6.36 Author Study sites Year Number of patients Cases Male/female (n/n) Age (years) 100a 300b 100 300 100 300 Han et al. [5] China 2015 260 90 83 56/34 49/34 60.23 ± 9.61 61.22 ± 13.38 Zhang et al. [7] China 2014 70 29 41 14/15 21/20 55 ± 12 56 ± 14 Zhang et al. [4] China 2015 166 78 88 52/26 56/32 58.7 ± 3.2 58.4 ± 2.6 Li et al. [6] China 2014 98 50 48 Nullc Nullc Nullc Nullc Wang et al. [3] China 2015 108 54 54 35/19 37/17 70.45 ± 5.31 71.44 ± 6.36 a 100, removing the CT <100 ml/24 h. b 300, removing the CT <300 ml/24 h. c Null for not mentioned. Table 2: General characteristics of included studies Author Study sites Year Number of patients Cases Male/female (n/n) Age (years) 100a 300b 100 300 100 300 Han et al. [5] China 2015 260 90 83 56/34 49/34 60.23 ± 9.61 61.22 ± 13.38 Zhang et al. [7] China 2014 70 29 41 14/15 21/20 55 ± 12 56 ± 14 Zhang et al. [4] China 2015 166 78 88 52/26 56/32 58.7 ± 3.2 58.4 ± 2.6 Li et al. [6] China 2014 98 50 48 Nullc Nullc Nullc Nullc Wang et al. [3] China 2015 108 54 54 35/19 37/17 70.45 ± 5.31 71.44 ± 6.36 Author Study sites Year Number of patients Cases Male/female (n/n) Age (years) 100a 300b 100 300 100 300 Han et al. [5] China 2015 260 90 83 56/34 49/34 60.23 ± 9.61 61.22 ± 13.38 Zhang et al. [7] China 2014 70 29 41 14/15 21/20 55 ± 12 56 ± 14 Zhang et al. [4] China 2015 166 78 88 52/26 56/32 58.7 ± 3.2 58.4 ± 2.6 Li et al. [6] China 2014 98 50 48 Nullc Nullc Nullc Nullc Wang et al. [3] China 2015 108 54 54 35/19 37/17 70.45 ± 5.31 71.44 ± 6.36 a 100, removing the CT <100 ml/24 h. b 300, removing the CT <300 ml/24 h. c Null for not mentioned. Risk of bias in included studies The risk of bias of those 5 trials is listed in Table 1. Only 4 trials mentioned adequate sequence generation clearly, and this can produce selective bias. Allocation concealment was not reported in the 5 trials, potentially leading to an unclear risk of selection bias, because it was possible for those responsible for recruiting the participants to alter their assignment if allocation was concealed. Blinding was reported in 2 trials and single blinding was used in those trials, but no detailed description about the blinding methods was mentioned. This might cause a high risk of performance bias or detection bias. Although a comprehensive literature search was conducted, some published and unpublished trials might have been missed, and this could potentially lead to nonpublication bias. Publication bias might exist. Meta-analysis results Amount of drainage in the first 24 h after operation The amount of drainage in the first 24 h after operation was reported in 3 trials. The fixed-effects model was used to perform meta-analysis because there was no significant heterogeneity between trials (I2 = 0%, P > 0.1). The meta-analysis result of this subject showed that the data included are not sufficient to show a significant difference between the 2 groups (MD = −20.44; 95% CI = −43.67–2.80, P = 0. 08) (Fig. 2). Figure 2: View largeDownload slide Forest plot of the amount of drainage in the first 24 h after operation. CI: confidence interval; SD: standard deviation. Figure 2: View largeDownload slide Forest plot of the amount of drainage in the first 24 h after operation. CI: confidence interval; SD: standard deviation. Duration of chest tube drainage The duration of CT drainage was reported in 4 trials. The random-effects model was used to perform meta-analysis because there was significant heterogeneity between trials (I2 = 96%, P < 0.1). Patients in the experimental group had a shorter duration of CT drainage, and data included showed that this difference was statistically significant between the 2 groups (MD = −44.07; 95% CI = −64.45 to −23.68, P-0.0001) (Fig. 3). Figure 3: View largeDownload slide Forest plot of chest tube duration after operation. CI: confidence interval; SD: standard deviation. Figure 3: View largeDownload slide Forest plot of chest tube duration after operation. CI: confidence interval; SD: standard deviation. The duration of CT drainage was also mentioned in Ye Zhang’s trials, but the author used the median and range to describe the duration. The original data were not received from the first author. Hospital stay after operation Hospital stay after operation was reported in 4 trials. The random-effects model was used to perform the meta-analysis because there was significant heterogeneity between trials (I2 = 93%, P-0.1). Patients in the experimental group had a shorter length of hospital stay after operations, and this difference was statistically significant between the 2 groups (MD = −0.20; 95% CI = −3.52 to −0.97, P = 0.0006) (Fig. 4). Figure 4: View largeDownload slide Forest plot of hospital stay after operation. CI: confidence interval; SD: standard deviation. Figure 4: View largeDownload slide Forest plot of hospital stay after operation. CI: confidence interval; SD: standard deviation. The hospital stay after operation was also mentioned in Ye Zhang’s trials, but the author used the median and range to describe the hospital stay. The original data were not received from the first author. Patient numbers of recurrent pleural fluid after removing chest tube The patient numbers of recurrent pleural fluid after removing the CT were reported in 3 trials. The fixed-effects model was used to perform the meta-analysis because there was no significant heterogeneity between trials (I2 = 0%, P > 0.1). The meta-analysis result of this subject showed that the data included are not sufficient to show a significant difference between the 2 groups (OR = 1.73; 95% CI = 0.74–4.07, P = 0. 21) (Fig. 5). Figure 5: View largeDownload slide Forest plot of the patient numbers of recurrent pleural fluid after removing the CT. CI: confidence interval. Figure 5: View largeDownload slide Forest plot of the patient numbers of recurrent pleural fluid after removing the CT. CI: confidence interval. Number of patients who need thoracentesis after removing chest tube The number of patients who need thoracentesis after removing the CT were reported in 3 trials. The fixed-effects model was used to perform the meta-analysis because there was no significant heterogeneity between the trials (I2 = 0%, P > 0.1). The meta-analysis result of the number of patients who need thoracentesis after removing the CT showed that the data included are not sufficient to show a significant difference between the 2 groups (OR = 1.73; 95% CI = 0.74–4.07, P = 0. 21) (Fig. 6). Figure 6: View largeDownload slide Forest plot of the number of patients who need thoracentesis after removing computed tomography. CI: confidence interval. Figure 6: View largeDownload slide Forest plot of the number of patients who need thoracentesis after removing computed tomography. CI: confidence interval. Patient numbers of atelectasis after operation The patient numbers of atelectasis after the operation were reported in 4 trials. The fixed-effects model was used to perform the meta-analysis because there was no significant heterogeneity between trials (I2 = 15%, P > 0.1). The meta-analysis result showed that the data included are not sufficient to show a significant difference between the 2 groups (risk ratio = 0.97; 95% CI = 0.52, 1.81, P = 0.93) (Supplementary Material, Fig. S1). DISCUSSION Summary of evidence This is a systematic review and meta-analysis of 5 RCTs to evaluate the efficiency and safety of using the threshold of daily drainage of 300 ml before removing the CTs. The present meta-analysis showed that increasing the daily threshold of 300 ml before removing the CT significantly decreased the duration of CT drainage and the hospital stay after the operation. In addition, there was no significant difference between the 2 groups on other meta-analytical end points, such as atelectasis after operation, need for thoracentesis after removing the CT and reaccumulation of pleural fluid after operation. The upper limits of the 95% CIs of ORs for recurrent pleural fluid (Fig. 5) and thoracentesis (Fig. 6) are larger than 4. This may be caused by an individual study of Zhang [7]. In the meta-analysis result of recurrent pleural fluid (Fig. 5), the incidence rate of events in the experimental group of Zhang’s study [7] was much higher than in the control group (4/41 comparing with 0/29). However, ‘1’ was included in the 95% CI of Zhang’s study [7] in this item, which means that the difference was not statistically significant between the 2 groups (OR = 7.08; 95% CI = 0.37–136.81). Beyond that, the weight of this study is 6.2%, much less than other included studies. Therefore, involving Zhang’s study would not significantly influence the consequence of the meta-analysis’s result of recurrent pleural fluid. For the same reason, the 95% CI of thoracentesis (Fig. 6) may also be influenced by Zhang’s study. The incidence rate of events in the experimental group is 2/41, and that in the control group is 0/29. Similarly, involving Zhang’s study would not significantly influence the consequence of the meta-analysis (OR = 3.73; 95% CI = 0.17–80.73; weight = 8.9%). After the lung resection, the placement of CTs is routine. Although the procedure is performed under optimal conditions to minimize risks, the morbidity rate is not low because of prolonged air leakage, empyema and other conditions. At the same time, CTs requiring specialized care in the hospital may incur consequent costs. Therefore, any reasonable reduction in the duration of CT drainage may decrease the likelihood of complications, length of hospital stays and hospitalization cost. The strategies for CT removal are still debatable and are based primarily on tradition and dogma more than the data [1]. The decision of pulling out the CT is generally based on the criterion that the fluid drainage is ≤100 ml for the least 24 h after lobectomy without air leakage or haemothorax [2]. Pleural fluid originates from the parietal pleura and is resorbed by the visceral pleura. Pleural effusion develops when the amount of fluid that enters the pleural space exceeds the amount that can be removed. With the re-expansion of the remaining lung after surgery, the pleural space is usually obliterated within several days to a week. In most instances, the fluid that remains is resorbed or becomes organized [7]. The wide use of staples, absorbable haemostatic sponge and fibrin glue in recent years has reduced the occurrence of air leakage, drainage time, postoperative hospitalization and pulmonary complication rate [9]. In recent years, several studies have been published, challenging the threshold of pleural drainage at which it is safe to remove the CT after pulmonary resection. Younes et al. [10] and Hessami et al. [11] had reported to remove the CT safely with a drainage ≤200 ml/day. McKenna et al. [12] has reported that pulling out CTs at 300 ml/day is feasible. Cerfolio and Bryant [13] summarized that it is viable to remove the CTs in a relatively higher threshold up to 450 ml/day according to their retrospective study. In addition, Göttgens et al. [14] found that CTs could be safely removed in the first 24–48 h after video-assisted thoracic surgery (bi-)lobectomy despite volumes of output of up to 400 ml/day. The volume threshold of the control group was selected based on the original standard for removing the CTs [2], and the volume threshold of the experimental group was selected based on the published RCTs that involve the comparison of various volume thresholds. The volume threshold of 300 ml/day had been studied in many RCTs before, which could provide relatively sufficient data for quantitative analysis. More aggressive volume threshold, such as 450 ml/day, was abandoned for lack of sufficient studies. Except for the duration of drainage and hospital stay, the meta-analysis result of the patient numbers of recurrent pleural fluid after removing the CT showed no significant difference between the 2 groups. After a lobectomy, the mechanical characteristics of the pleural space are altered by thoracic surgery conspicuously even after closure of the chest. Furthermore, the increase in microvascular filtration, pulmonary microvascular pressure and pleural membrane permeability combined with the absence of an efficient lymphatic pleural drainage may cause the increased pleural effusion accumulation [1]. To avoid lung overdistension, appropriate fluid and gas volume remained in the pleural cavity are significantly meaningful in the early postoperative period. However, Xie et al. [15] mentioned that excess fluid may hinder the expansion of remanding lung, which may further increase the fluid accumulation. Therefore, setting a suitable threshold may decrease the chance of fluid reaccumulation. The data from the trial of Xie et al. [15] showed that patients with daily drainage increasing to 450 ml/day before removing the CTs were more likely to have recurrent pleural fluid. A higher volume threshold may also increase the risk of thoracentesis compared with the volume threshold of 300 ml/day. The factors above indicate that, compared with 100 ml/day, the application of 300 ml/day as a volume threshold could significantly decrease the duration of CT placement and hospital stay after operation. In addition, 300 ml/day may be safer than 450 ml/day in preventing reaccumulation of pleural effusion. The main finding of our meta-analysis seems to be consistent with a previous RCT performed by Xie et al. [15], which suggested that, compared with 150 ml/day, the volume threshold of 300 ml/day for CT removal can shorten the postoperative hospital stay volume threshold. This trial also summarized that the volume threshold of 300 ml/day can decrease the postoperative pain of patients. This item was not involved in the present meta-analysis. Three of the 5 RCTs refer to pain management and visual analogue scale of pain grade, but measuring times and description methods of this item is different among those 3 RCTs. The amount of fluid is the primary factor considered that influences the duration of CT in this study. Several factors such as quality of the fluid, number of CTs, CT diameter and general status of the patients may also influence the CT management strategy. For the purpose of reducing the heterogeneity of the included trails, this study had only taken trails aiming at volume threshold into consideration. Further studies and trails should focus on those subjects. To the best of our knowledge, no published meta-analysis has focused on the volume threshold before removing the CTs after lobectomy. As reported in this review, the majority of the studies employed small sample sizes and lacked the statistical power necessary to make a clear statement regarding the utility of the volume threshold of 300 ml/day. A meta-analysis, such as that performed in this system review, is a potentially useful tool in this context because pooling data can result in a more powerful conclusion, as opposed to the results obtained from smaller individual trials. Limitations Limitations and implications for RCTs and meta-analysis in this review are listed as follows: some limitations should be considered, and some improvements should be made for further studies. In 5 trials, the type and number of CTs and the method of the pulmonary lobectomy are different; the meta-analysis result may be influenced. This review is limited by the methodological quality of the included RCTs. Some literatures from local journals were included. Some information is insufficient to permit the defined judgement, although every attempt was made to contact the authors of the included trials through e-mail or telephone. It is also recommended that additional RCTs are performed and reported according to the recommendations of Cochrane Library, offering a standard way to improve the quality of research. Only English and Chinese language literature articles were considered for inclusion. If the search had been extended to include literature published in other languages, then additional relevant trials would have been identified. Currently, there are few data comparing the different volume threshold applications in patients after pulmonary lobectomy. Therefore, additional RCTs should be conducted in these specific subgroups, especially a volume threshold of higher than 300 ml/day. This study has a limitation because of its sample size. We could not identify the effect modifiers that may be attributed to the low statistical power. The number of CTs being used in patients after operations is different: 1 trial [5] used only 1 CT in both groups, 1 trial [4] used 1 CT in the experimental group and 2 CTs in the control group, 2 trials [3, 6] used 2 CTs when the upper lung was resected and 1 CT was used in other type of lobectomies. One trial mentioned that both algorithms might be used in 2 groups [7]. The meta-analysis result may be influenced. The trials included in this analysis were all preformed in China within 3 years. Although all the RCTs involved came from 1 country, some non-randomized trials or retrospective studies in other countries [9–14] had come up with similar conclusions. The usage of relatively aggressive drawing strategies other than 100 ml/day is recommended after looking back to those trails from other countries. As a result, the conclusion of this study may also be helpful for clinical workers of other countries. Besides that, this coincidence may lead to smaller heterogeneity in the surgical technique and genomic differences. We still need further high-quality, multicentre, randomized, controlled trials from other countries and regions. The postoperative complications that were analysed in this review only included reaccumulation of pleural fluid and atelectasis that was caused by the limited data from current studies. These may not be sufficient to evaluate the overall effects of 2 drainage strategies. Further studies should focus on more complications and items that contribute to the assessment of those 2 strategies. The accumulated pleural fluid after CT removal was mentioned in 4 trails [4–7]. References [6, 7] recorded this outcome from CT removal to 1 week after discharge, Ref. [5] recorded the outcome in 1 week after CT removal, and Ref. [4] had not mentioned the period of pleural fluid reaccumulation. The short follow-up and difference in observation time may influence the meta-analysis result. Further trials should observe this outcome for 1 month at least after CT removal to verify the correlation between CT management and pleural fluid reaccumulation [1]. CONCLUSIONS The meta-analysis result showed that increasing the threshold of daily drainage to 300 ml before removing the CT showed the same clinically meaningful and statistically improvement in amount of drainage in the first 24 h after operation, the patient numbers of pleural fluid reaccumulation, the patient numbers of atelectasis after operation and the number of patients requiring thoracentesis after removing the CT; however, using the volume threshold of 300 ml significantly decreases the duration of CT drainage and the hospital stay after operation compared with patients removing the CT when the daily drainage is <100 ml. Hence, patients using the daily threshold of 300 ml before removing the CT may be more suitable for patients after a pulmonary lobectomy. Although there is convincing evidence to confirm the results mentioned herein, they still need to be confirmed by large-sample, multicentre, randomized, controlled trials. SUPPLEMENTARY MATERIAL Supplementary material is available at ICVTS online. Funding This study was supported by Beijing Municipal Administration of Hospitals Ascent Plan [DFL20151501]. Conflict of interest: none declared. REFERENCES 1 Brunelli A , Beretta E , Cassivi SD , Cerfolio RJ , Detterbeck F , Kiefer T et al. Consensus definitions to promote an evidence-based approach to management of the pleural space. A collaborative proposal by ESTS, AATS, STS, and GTSC . Eur J Cardiothorac Surg 2011 ; 40 : 291 – 7 . Google Scholar CrossRef Search ADS PubMed 2 Gomez-Caro A , Roca MJ , Torres J , Cascales P , Terol E , Castaner J et al. Successful use of a single chest drain postlobectomy instead of two classical drains: a randomized study . Eur J Cardiothorac Surg 2006 ; 29 : 562 – 6 . Google Scholar CrossRef Search ADS PubMed 3 Wang J , Ni B , Ma H. The application of fast-track surgery in thoracoscopic lung cancer operation for the aged patients . J Laparosc Surg 2015 ; 8 : 581 – 5 . 4 Zhang W , Liu K , Pei Y , Zhao J. Improved chest tube application promotes fast track recovery after lobectomy . Chin J Front Med Sci (Elect Ver) 2015 ; 7 : 56 – 8 . 5 Han H , Zhang X , Wang D , Yao P. The association between drainage volume and removal of chest tube after video-assisted thoracoscopic lobectomy . Tianjin Med J 2015 ; 1 : 85 – 7 . 6 Li W , Jin Z , Duan DK. Comparison of the timing of thoracic drainage tube removal in gerontal patients after lung cancer surgery . Chin J Gerontol 2014 ; 16 : 4534 – 6 . 7 Zhang Y , Li H , Hu B , Li T , Miao J-B , You B et al. A prospective randomized single-blind control study of volume threshold for chest tube removal following lobectomy . World J Surg 2014 ; 38 : 60 – 7 . Google Scholar CrossRef Search ADS PubMed 8 Teng YJ , Pan SM , Liu YL , Yang KH , Zhang YC , Tian JH et al. A meta-analysis of randomized controlled trials of fixation versus nonfixation of mesh in laparoscopic total extraperitoneal inguinal hernia repair . Surg Endosc 2011 ; 25 : 2849 – 58 . Google Scholar CrossRef Search ADS PubMed 9 Bjerregaard LS , Jensen K , Petersen RH , Hansen HJ. Early chest tube removal after video-assisted thoracic surgery lobectomy with serous fluid production up to 500 ml/day . Eur J Cardiothorac Surg 2014 ; 45 : 241 – 6 . Google Scholar CrossRef Search ADS PubMed 10 Younes RN , Gross JL , Aguiar S , Haddad FJ , Deheinzelin D. When to remove a chest tube? A randomized study with subsequent prospective consecutive validation . J Am Coll Surg 2002 ; 195 : 658 – 62 . Google Scholar CrossRef Search ADS PubMed 11 Hessami MA , Najafi F , Hatami S. Volume threshold for chest tube removal: a randomized controlled trial . J Inj Violence Res 2009 ; 1 : 33 – 6 . Google Scholar CrossRef Search ADS PubMed 12 McKenna RJ , Mahtabifard A , Pickens A , Kusuanco D , Fuller CB. Fast-tracking after video-assisted thoracoscopic surgery lobectomy, segmentectomy, and pneumonectomy . Ann Thorac Surg 2007 ; 84 : 1663 – 7; discussion 1667–8. Google Scholar CrossRef Search ADS PubMed 13 Cerfolio RJ , Bryant AS. Results of a prospective algorithm to remove chest tubes after pulmonary resection with high output . J Thorac Cardiovasc Surg 2008 ; 135 : 269 – 73 . Google Scholar CrossRef Search ADS PubMed 14 Göttgens KW , Siebenga J , Belgers EH , van Huijstee PJ , Bollen EC. Early removal of the chest tube after complete video-assisted thoracoscopic lobectomies . Eur J Cardiothorac Surg 2011 ; 39 : 575 – 8 . Google Scholar CrossRef Search ADS PubMed 15 Xie HY , Xu K , Tang JX , Bian W , Ma HT , Zhao J et al. A prospective randomized, controlled trial deems a drainage of 300 ml/day safe before removal of the last chest drain after video-assisted thoracoscopic surgery lobectomy . Interact CardioVasc Thorac Surg 2015 ; 21 : 200 – 5 . Google Scholar CrossRef Search ADS PubMed © The Author(s) 2018. Published by Oxford University Press on behalf of the European Association for Cardio-Thoracic Surgery. All rights reserved. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices) http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Interactive CardioVascular and Thoracic Surgery Oxford University Press

The volume threshold of 300 versus 100 ml/day for chest tube removal after pulmonary lobectomy: a meta-analysis

Loading next page...
 
/lp/ou_press/the-volume-threshold-of-300-versus-100-ml-day-for-chest-tube-removal-io6JI55nEo
Publisher
Oxford University Press
Copyright
© The Author(s) 2018. Published by Oxford University Press on behalf of the European Association for Cardio-Thoracic Surgery. All rights reserved.
ISSN
1569-9293
eISSN
1569-9285
D.O.I.
10.1093/icvts/ivy150
Publisher site
See Article on Publisher Site

Abstract

Abstract OBJECTIVES In this meta-analysis, we conducted a pooled analysis of clinical studies comparing the efficacy of a volume threshold of 300 ml/day before removing a chest tube (CT) versus 100 ml/day after a lobectomy. METHODS According to the recommendations of the Cochrane Collaboration, we established a rigorous study protocol. We performed a systematic electronic search of PubMed, Embase, Cochrane Library, Web of Science databases, CNKI, the Wanfang database, CBMdisc and Google Scholar to identify articles to include in our meta-analysis. A literature search was performed using relevant keywords. A meta-analysis was performed using RevMan© software. RESULTS Five studies, published between 2014 and 2015, including 615 patients (314 patients who had the CT removed when daily drainage was <300 ml and 301 patients who had the CT removed when daily drainage was <100 ml) met the selection criteria. From the available data, the patients using the volume threshold of 300 ml/day had a significantly decreased duration of drainage [MD = −44.07; 95% confidence interval (CI) −64.45 to −23.68; P < 0.0001] and hospital stay after operation (MD = −2.25; 95% CI −3.52 to −0.97; P = 0.0006) compared with patients using a volume threshold of 100 ml/day after a pulmonary lobectomy. However, no significant differences were observed in postoperative complications, such as pleural fluid reaccumulation [Odds ratio (OR) = 1.73; 95% CI = 0.74–4.07; P = 0.21] and atelectasis (OR = 0.97; 95% CI = 0.52–1.81; P = 0.93). Thoracentesis rates after removing the CT also showed no significant difference (OR = 1.53; 95% CI 0.55–4.22; P = 0.41). CONCLUSIONS Our results showed that a higher volume threshold, up to 300 ml/day, is effective in reducing hospitalization times and duration of drainage in patients who undergo a lobectomy. Moreover, the volume threshold of 300 ml/day does not increase the occurrence of postoperative atelectasis, pleural fluid reaccumulation and thoracentesis rates. However, this review is limited by the methodological quality of the included trials, and additional studies according to the recommendations of Cochrane Library are appreciated. Chest tube, Lobectomy, Drainage, Volume threshold, Meta-analysis INTRODUCTION The need for chest tube (CT) drainage is common after operations on the pleural space or chest trauma. In general, the use of CT could help expend postoperative residual lungs and drain the fluid and air from the pleural space. In addition, CT could also be an indication of prolonged air leaks, haemothorax, empyema and chylothorax. CTs are known to have a long history of clinical use; however, indications for its removal are often controversial, including parameters such as volume threshold, drainage time and protein content of the pleural fluid [1]. There is wide variation in the mean duration of CT drainage among several institutions. In general, the standard for removing the tube requires no air leakage or haemothorax and fluid drainage of <100 ml for at least 24 h after a lobectomy [2]. Recently, several studies have reported the safe removal of the CT with drainage exceeding 100 ml for at least 24 h, and the new volume threshold of 300 ml/day has been used in many randomized trials [3–7]. Although some recent randomized trials have compared the effectiveness of using a volume threshold of 300 or 100 ml/day before removing the CTs, few available data could support which algorithm is more effective. There are no evidence-based consensus recommendations for the optimal volume threshold to be used. The main objective of this meta-analysis is to conduct a pooled analysis of clinical studies to compare the volume threshold of 300 ml/day with 100 ml/day before removing the CT with a statistical power much higher than that in each trial. Objective outcomes such as duration of drainage, volume of drainage, postoperative hospital stay, pleural fluid reaccumulation, thoracentesis rate and complications after pulmonary lobectomy were collected and analysed. PATIENTS AND METHODS A rigorous study protocol was established according to the recommendations of the Cochrane Collaboration. Inclusion criteria Types of studies All published RCTs comparing volume thresholds of 100 and 300 ml/day before removing a CT after pulmonary lobectomy were considered eligible for the meta-analysis. Language was limited to English and Chinese. Abstracts or unpublished data were included only if sufficient information on interventions and outcomes was available and if the final results were confirmed by contact with the first author. Types of participants The following RCTs were included: Trials comprising patients who were scheduled to undergo the pulmonary lobectomy or bilobectomy operation. Trials comparing volume thresholds of 100 and 300 ml/day before CT removal after pulmonary lobectomy. Trials in which the primary outcome measure was the duration of drainage or the amount of drainage. The secondary outcome measures were hospital stay after operation, postoperative complications or the number of patients who need thoracentesis after removing the CT. Exclusion criteria Patients who suffer from postoperative prolonged air leakage. Patients who suffer from densely bloody, purulent or cloudy pleural effusion after operations. Patients who were expected to have increased postoperative haemorrhage such as haematological system diseases and pleural extensive adhesion. Patients who suffer from adverse cardiovascular and cerebrovascular events. Literature search The Cochrane Central Register of Controlled Trials in the PubMed, Embase, Cochrane Library, Web of Science databases, CNKI, Wanfang database and CBMdisc were searched for RCTs comparing different volume thresholds before removing CTs after pulmonary lobectomy without language restriction. Moreover, Google Scholar, Baidu Scholar and reference lists of all the included studies were searched for additional reports. Contact with the authors was initiated by e-mail or telephone if any information was not available. The search strategies used the following major terms: ‘Pneumonectomy[Mesh]’, ‘pulmonary lobectomy[Mesh]’, ‘bilobectomy’, ‘Chest Tubes[Mesh]’, ‘Drainage[Mesh]’, ‘Randomized Controlled Trial OR random’, ‘Pneumonectomy OR Lung Volume Reduction OR pulmonary lobectomy OR Lobectomy OR bilobectomy OR pulmonary resection OR lung resection’ and ‘chest drain OR pleural drainage OR thoracic drainage OR Chest Tube’. Data extraction and quality assessment The described search strategy was used to obtain titles and abstracts of RCTs that were relevant to this review. Two reviewers independently assessed the titles and abstracts of all identified trials to confirm the fulfilment of inclusion criteria, and the duration of drainage, volume of drainage in the first 24 h, postoperative hospital stay, pleural fluid reaccumulation, thoracentesis rate and complications after pulmonary lobectomy were collected; data abstraction was performed independently by 2 reviewers. Any difference of opinion or disagreement that arose in the course of searching, data abstraction, quality assessment or other related work between the 2 investigators was resolved by discussion. Quality assessment was performed after full-text reading, and trials of low quality would be excluded. The risk of bias of the included RCTs was assessed according to Cochrane Handbook 5.1.0 and the 5 criteria used were listed in Table 1. Each entry was definitively judged by an answer (yes/no/unclear), which stands for a low, a high and an unclear or unknown risk of bias [8]. Table 1: Risk of bias in included studies Author Adequate sequence generation Allocation concealment Blinding Incomplete outcome data Free of selective reporting Free of other bias Han et al. [5] Yes (table of random numbers) Unclear Unclear Yes Unclear Unclear Zhang et al. [7] Yes (table of random numbers) Unclear Unclear Yes Unclear Unclear Zhang et al. [4] Yes (table of random numbers) Unclear Unclear Yes Unclear Unclear Li et al. [6] Unclear Unclear Unclear Yes Unclear Unclear Wang et al. [3] Yes (table of random numbers) Unclear Unclear Yes Unclear Unclear Author Adequate sequence generation Allocation concealment Blinding Incomplete outcome data Free of selective reporting Free of other bias Han et al. [5] Yes (table of random numbers) Unclear Unclear Yes Unclear Unclear Zhang et al. [7] Yes (table of random numbers) Unclear Unclear Yes Unclear Unclear Zhang et al. [4] Yes (table of random numbers) Unclear Unclear Yes Unclear Unclear Li et al. [6] Unclear Unclear Unclear Yes Unclear Unclear Wang et al. [3] Yes (table of random numbers) Unclear Unclear Yes Unclear Unclear Table 1: Risk of bias in included studies Author Adequate sequence generation Allocation concealment Blinding Incomplete outcome data Free of selective reporting Free of other bias Han et al. [5] Yes (table of random numbers) Unclear Unclear Yes Unclear Unclear Zhang et al. [7] Yes (table of random numbers) Unclear Unclear Yes Unclear Unclear Zhang et al. [4] Yes (table of random numbers) Unclear Unclear Yes Unclear Unclear Li et al. [6] Unclear Unclear Unclear Yes Unclear Unclear Wang et al. [3] Yes (table of random numbers) Unclear Unclear Yes Unclear Unclear Author Adequate sequence generation Allocation concealment Blinding Incomplete outcome data Free of selective reporting Free of other bias Han et al. [5] Yes (table of random numbers) Unclear Unclear Yes Unclear Unclear Zhang et al. [7] Yes (table of random numbers) Unclear Unclear Yes Unclear Unclear Zhang et al. [4] Yes (table of random numbers) Unclear Unclear Yes Unclear Unclear Li et al. [6] Unclear Unclear Unclear Yes Unclear Unclear Wang et al. [3] Yes (table of random numbers) Unclear Unclear Yes Unclear Unclear Statistical analysis Statistical analysis for the meta-analysis was performed with RevMan (Review Manager Version 5.3.3 for windows; Cochrane Collaboration, Oxford, UK, 2014). A pooled relative risk was calculated with 95% confidence intervals (CIs). Odds ratios (ORs) or risk ratios were used for dichotomous variables, and a mean difference was used for continuous variables. χ2 heterogeneity tests with statistical significance at P < 0.10 were applied to test statistical heterogeneity among trials. The I2 statistics were also applied to assess the proportion of variability in the results attributable to heterogeneity across studies, with I2 < 50% considered as low-level heterogeneity and I2 > 50% as high-level heterogeneity. Fixed-effects model was used if there was no statistically significant heterogeneity (P ≥ 0.10, I2 < 50%), and the random-effects model was used if there was significant heterogeneity (P < 0.10, I2 > 50%). Because the characteristics of the patients, operation methods, number of CTs used after an operation and other confounding factors were not consistent among studies, we further conducted sensitivity analyses to explore possible explanations for heterogeneity. Sensitivity analysis was carried out by omitting 1 trial in each turn. The sensitivity analysis did not materially alter the pooled results, adding robustness to our findings. Intention-to-treat analysis was not performed because of insufficient information about loss to follow-up in treatment and control groups. A funnel plot was not used because of the limited number of RCTs. RESULTS Study description The flowchart for literature screening is presented in Fig. 1. According to the established search strategy that was used, a total of 838 potentially relevant literature items were identified in the databases, and 27 duplicate studies were excluded by NoteExpress software. After screening the titles and abstracts of the remaining 791 studies, 659 irrelevant studies were excluded, leaving 132 studies for further assessment. After the full-text review of the 132 studies, we excluded 127 of them. Therefore, 5 trials that fulfilled the inclusion criteria were included [8]. Figure 1: View largeDownload slide Flow diagram: selection and evaluation process of the eligible studies. Figure 1: View largeDownload slide Flow diagram: selection and evaluation process of the eligible studies. Characteristics of the included studies The characteristics of the included studies are presented in Table 2; all 5 trials were performed in China. Both groups were well matched at baseline from the information on all trials. Table 2: General characteristics of included studies Author Study sites Year Number of patients Cases Male/female (n/n) Age (years) 100a 300b 100 300 100 300 Han et al. [5] China 2015 260 90 83 56/34 49/34 60.23 ± 9.61 61.22 ± 13.38 Zhang et al. [7] China 2014 70 29 41 14/15 21/20 55 ± 12 56 ± 14 Zhang et al. [4] China 2015 166 78 88 52/26 56/32 58.7 ± 3.2 58.4 ± 2.6 Li et al. [6] China 2014 98 50 48 Nullc Nullc Nullc Nullc Wang et al. [3] China 2015 108 54 54 35/19 37/17 70.45 ± 5.31 71.44 ± 6.36 Author Study sites Year Number of patients Cases Male/female (n/n) Age (years) 100a 300b 100 300 100 300 Han et al. [5] China 2015 260 90 83 56/34 49/34 60.23 ± 9.61 61.22 ± 13.38 Zhang et al. [7] China 2014 70 29 41 14/15 21/20 55 ± 12 56 ± 14 Zhang et al. [4] China 2015 166 78 88 52/26 56/32 58.7 ± 3.2 58.4 ± 2.6 Li et al. [6] China 2014 98 50 48 Nullc Nullc Nullc Nullc Wang et al. [3] China 2015 108 54 54 35/19 37/17 70.45 ± 5.31 71.44 ± 6.36 a 100, removing the CT <100 ml/24 h. b 300, removing the CT <300 ml/24 h. c Null for not mentioned. Table 2: General characteristics of included studies Author Study sites Year Number of patients Cases Male/female (n/n) Age (years) 100a 300b 100 300 100 300 Han et al. [5] China 2015 260 90 83 56/34 49/34 60.23 ± 9.61 61.22 ± 13.38 Zhang et al. [7] China 2014 70 29 41 14/15 21/20 55 ± 12 56 ± 14 Zhang et al. [4] China 2015 166 78 88 52/26 56/32 58.7 ± 3.2 58.4 ± 2.6 Li et al. [6] China 2014 98 50 48 Nullc Nullc Nullc Nullc Wang et al. [3] China 2015 108 54 54 35/19 37/17 70.45 ± 5.31 71.44 ± 6.36 Author Study sites Year Number of patients Cases Male/female (n/n) Age (years) 100a 300b 100 300 100 300 Han et al. [5] China 2015 260 90 83 56/34 49/34 60.23 ± 9.61 61.22 ± 13.38 Zhang et al. [7] China 2014 70 29 41 14/15 21/20 55 ± 12 56 ± 14 Zhang et al. [4] China 2015 166 78 88 52/26 56/32 58.7 ± 3.2 58.4 ± 2.6 Li et al. [6] China 2014 98 50 48 Nullc Nullc Nullc Nullc Wang et al. [3] China 2015 108 54 54 35/19 37/17 70.45 ± 5.31 71.44 ± 6.36 a 100, removing the CT <100 ml/24 h. b 300, removing the CT <300 ml/24 h. c Null for not mentioned. Risk of bias in included studies The risk of bias of those 5 trials is listed in Table 1. Only 4 trials mentioned adequate sequence generation clearly, and this can produce selective bias. Allocation concealment was not reported in the 5 trials, potentially leading to an unclear risk of selection bias, because it was possible for those responsible for recruiting the participants to alter their assignment if allocation was concealed. Blinding was reported in 2 trials and single blinding was used in those trials, but no detailed description about the blinding methods was mentioned. This might cause a high risk of performance bias or detection bias. Although a comprehensive literature search was conducted, some published and unpublished trials might have been missed, and this could potentially lead to nonpublication bias. Publication bias might exist. Meta-analysis results Amount of drainage in the first 24 h after operation The amount of drainage in the first 24 h after operation was reported in 3 trials. The fixed-effects model was used to perform meta-analysis because there was no significant heterogeneity between trials (I2 = 0%, P > 0.1). The meta-analysis result of this subject showed that the data included are not sufficient to show a significant difference between the 2 groups (MD = −20.44; 95% CI = −43.67–2.80, P = 0. 08) (Fig. 2). Figure 2: View largeDownload slide Forest plot of the amount of drainage in the first 24 h after operation. CI: confidence interval; SD: standard deviation. Figure 2: View largeDownload slide Forest plot of the amount of drainage in the first 24 h after operation. CI: confidence interval; SD: standard deviation. Duration of chest tube drainage The duration of CT drainage was reported in 4 trials. The random-effects model was used to perform meta-analysis because there was significant heterogeneity between trials (I2 = 96%, P < 0.1). Patients in the experimental group had a shorter duration of CT drainage, and data included showed that this difference was statistically significant between the 2 groups (MD = −44.07; 95% CI = −64.45 to −23.68, P-0.0001) (Fig. 3). Figure 3: View largeDownload slide Forest plot of chest tube duration after operation. CI: confidence interval; SD: standard deviation. Figure 3: View largeDownload slide Forest plot of chest tube duration after operation. CI: confidence interval; SD: standard deviation. The duration of CT drainage was also mentioned in Ye Zhang’s trials, but the author used the median and range to describe the duration. The original data were not received from the first author. Hospital stay after operation Hospital stay after operation was reported in 4 trials. The random-effects model was used to perform the meta-analysis because there was significant heterogeneity between trials (I2 = 93%, P-0.1). Patients in the experimental group had a shorter length of hospital stay after operations, and this difference was statistically significant between the 2 groups (MD = −0.20; 95% CI = −3.52 to −0.97, P = 0.0006) (Fig. 4). Figure 4: View largeDownload slide Forest plot of hospital stay after operation. CI: confidence interval; SD: standard deviation. Figure 4: View largeDownload slide Forest plot of hospital stay after operation. CI: confidence interval; SD: standard deviation. The hospital stay after operation was also mentioned in Ye Zhang’s trials, but the author used the median and range to describe the hospital stay. The original data were not received from the first author. Patient numbers of recurrent pleural fluid after removing chest tube The patient numbers of recurrent pleural fluid after removing the CT were reported in 3 trials. The fixed-effects model was used to perform the meta-analysis because there was no significant heterogeneity between trials (I2 = 0%, P > 0.1). The meta-analysis result of this subject showed that the data included are not sufficient to show a significant difference between the 2 groups (OR = 1.73; 95% CI = 0.74–4.07, P = 0. 21) (Fig. 5). Figure 5: View largeDownload slide Forest plot of the patient numbers of recurrent pleural fluid after removing the CT. CI: confidence interval. Figure 5: View largeDownload slide Forest plot of the patient numbers of recurrent pleural fluid after removing the CT. CI: confidence interval. Number of patients who need thoracentesis after removing chest tube The number of patients who need thoracentesis after removing the CT were reported in 3 trials. The fixed-effects model was used to perform the meta-analysis because there was no significant heterogeneity between the trials (I2 = 0%, P > 0.1). The meta-analysis result of the number of patients who need thoracentesis after removing the CT showed that the data included are not sufficient to show a significant difference between the 2 groups (OR = 1.73; 95% CI = 0.74–4.07, P = 0. 21) (Fig. 6). Figure 6: View largeDownload slide Forest plot of the number of patients who need thoracentesis after removing computed tomography. CI: confidence interval. Figure 6: View largeDownload slide Forest plot of the number of patients who need thoracentesis after removing computed tomography. CI: confidence interval. Patient numbers of atelectasis after operation The patient numbers of atelectasis after the operation were reported in 4 trials. The fixed-effects model was used to perform the meta-analysis because there was no significant heterogeneity between trials (I2 = 15%, P > 0.1). The meta-analysis result showed that the data included are not sufficient to show a significant difference between the 2 groups (risk ratio = 0.97; 95% CI = 0.52, 1.81, P = 0.93) (Supplementary Material, Fig. S1). DISCUSSION Summary of evidence This is a systematic review and meta-analysis of 5 RCTs to evaluate the efficiency and safety of using the threshold of daily drainage of 300 ml before removing the CTs. The present meta-analysis showed that increasing the daily threshold of 300 ml before removing the CT significantly decreased the duration of CT drainage and the hospital stay after the operation. In addition, there was no significant difference between the 2 groups on other meta-analytical end points, such as atelectasis after operation, need for thoracentesis after removing the CT and reaccumulation of pleural fluid after operation. The upper limits of the 95% CIs of ORs for recurrent pleural fluid (Fig. 5) and thoracentesis (Fig. 6) are larger than 4. This may be caused by an individual study of Zhang [7]. In the meta-analysis result of recurrent pleural fluid (Fig. 5), the incidence rate of events in the experimental group of Zhang’s study [7] was much higher than in the control group (4/41 comparing with 0/29). However, ‘1’ was included in the 95% CI of Zhang’s study [7] in this item, which means that the difference was not statistically significant between the 2 groups (OR = 7.08; 95% CI = 0.37–136.81). Beyond that, the weight of this study is 6.2%, much less than other included studies. Therefore, involving Zhang’s study would not significantly influence the consequence of the meta-analysis’s result of recurrent pleural fluid. For the same reason, the 95% CI of thoracentesis (Fig. 6) may also be influenced by Zhang’s study. The incidence rate of events in the experimental group is 2/41, and that in the control group is 0/29. Similarly, involving Zhang’s study would not significantly influence the consequence of the meta-analysis (OR = 3.73; 95% CI = 0.17–80.73; weight = 8.9%). After the lung resection, the placement of CTs is routine. Although the procedure is performed under optimal conditions to minimize risks, the morbidity rate is not low because of prolonged air leakage, empyema and other conditions. At the same time, CTs requiring specialized care in the hospital may incur consequent costs. Therefore, any reasonable reduction in the duration of CT drainage may decrease the likelihood of complications, length of hospital stays and hospitalization cost. The strategies for CT removal are still debatable and are based primarily on tradition and dogma more than the data [1]. The decision of pulling out the CT is generally based on the criterion that the fluid drainage is ≤100 ml for the least 24 h after lobectomy without air leakage or haemothorax [2]. Pleural fluid originates from the parietal pleura and is resorbed by the visceral pleura. Pleural effusion develops when the amount of fluid that enters the pleural space exceeds the amount that can be removed. With the re-expansion of the remaining lung after surgery, the pleural space is usually obliterated within several days to a week. In most instances, the fluid that remains is resorbed or becomes organized [7]. The wide use of staples, absorbable haemostatic sponge and fibrin glue in recent years has reduced the occurrence of air leakage, drainage time, postoperative hospitalization and pulmonary complication rate [9]. In recent years, several studies have been published, challenging the threshold of pleural drainage at which it is safe to remove the CT after pulmonary resection. Younes et al. [10] and Hessami et al. [11] had reported to remove the CT safely with a drainage ≤200 ml/day. McKenna et al. [12] has reported that pulling out CTs at 300 ml/day is feasible. Cerfolio and Bryant [13] summarized that it is viable to remove the CTs in a relatively higher threshold up to 450 ml/day according to their retrospective study. In addition, Göttgens et al. [14] found that CTs could be safely removed in the first 24–48 h after video-assisted thoracic surgery (bi-)lobectomy despite volumes of output of up to 400 ml/day. The volume threshold of the control group was selected based on the original standard for removing the CTs [2], and the volume threshold of the experimental group was selected based on the published RCTs that involve the comparison of various volume thresholds. The volume threshold of 300 ml/day had been studied in many RCTs before, which could provide relatively sufficient data for quantitative analysis. More aggressive volume threshold, such as 450 ml/day, was abandoned for lack of sufficient studies. Except for the duration of drainage and hospital stay, the meta-analysis result of the patient numbers of recurrent pleural fluid after removing the CT showed no significant difference between the 2 groups. After a lobectomy, the mechanical characteristics of the pleural space are altered by thoracic surgery conspicuously even after closure of the chest. Furthermore, the increase in microvascular filtration, pulmonary microvascular pressure and pleural membrane permeability combined with the absence of an efficient lymphatic pleural drainage may cause the increased pleural effusion accumulation [1]. To avoid lung overdistension, appropriate fluid and gas volume remained in the pleural cavity are significantly meaningful in the early postoperative period. However, Xie et al. [15] mentioned that excess fluid may hinder the expansion of remanding lung, which may further increase the fluid accumulation. Therefore, setting a suitable threshold may decrease the chance of fluid reaccumulation. The data from the trial of Xie et al. [15] showed that patients with daily drainage increasing to 450 ml/day before removing the CTs were more likely to have recurrent pleural fluid. A higher volume threshold may also increase the risk of thoracentesis compared with the volume threshold of 300 ml/day. The factors above indicate that, compared with 100 ml/day, the application of 300 ml/day as a volume threshold could significantly decrease the duration of CT placement and hospital stay after operation. In addition, 300 ml/day may be safer than 450 ml/day in preventing reaccumulation of pleural effusion. The main finding of our meta-analysis seems to be consistent with a previous RCT performed by Xie et al. [15], which suggested that, compared with 150 ml/day, the volume threshold of 300 ml/day for CT removal can shorten the postoperative hospital stay volume threshold. This trial also summarized that the volume threshold of 300 ml/day can decrease the postoperative pain of patients. This item was not involved in the present meta-analysis. Three of the 5 RCTs refer to pain management and visual analogue scale of pain grade, but measuring times and description methods of this item is different among those 3 RCTs. The amount of fluid is the primary factor considered that influences the duration of CT in this study. Several factors such as quality of the fluid, number of CTs, CT diameter and general status of the patients may also influence the CT management strategy. For the purpose of reducing the heterogeneity of the included trails, this study had only taken trails aiming at volume threshold into consideration. Further studies and trails should focus on those subjects. To the best of our knowledge, no published meta-analysis has focused on the volume threshold before removing the CTs after lobectomy. As reported in this review, the majority of the studies employed small sample sizes and lacked the statistical power necessary to make a clear statement regarding the utility of the volume threshold of 300 ml/day. A meta-analysis, such as that performed in this system review, is a potentially useful tool in this context because pooling data can result in a more powerful conclusion, as opposed to the results obtained from smaller individual trials. Limitations Limitations and implications for RCTs and meta-analysis in this review are listed as follows: some limitations should be considered, and some improvements should be made for further studies. In 5 trials, the type and number of CTs and the method of the pulmonary lobectomy are different; the meta-analysis result may be influenced. This review is limited by the methodological quality of the included RCTs. Some literatures from local journals were included. Some information is insufficient to permit the defined judgement, although every attempt was made to contact the authors of the included trials through e-mail or telephone. It is also recommended that additional RCTs are performed and reported according to the recommendations of Cochrane Library, offering a standard way to improve the quality of research. Only English and Chinese language literature articles were considered for inclusion. If the search had been extended to include literature published in other languages, then additional relevant trials would have been identified. Currently, there are few data comparing the different volume threshold applications in patients after pulmonary lobectomy. Therefore, additional RCTs should be conducted in these specific subgroups, especially a volume threshold of higher than 300 ml/day. This study has a limitation because of its sample size. We could not identify the effect modifiers that may be attributed to the low statistical power. The number of CTs being used in patients after operations is different: 1 trial [5] used only 1 CT in both groups, 1 trial [4] used 1 CT in the experimental group and 2 CTs in the control group, 2 trials [3, 6] used 2 CTs when the upper lung was resected and 1 CT was used in other type of lobectomies. One trial mentioned that both algorithms might be used in 2 groups [7]. The meta-analysis result may be influenced. The trials included in this analysis were all preformed in China within 3 years. Although all the RCTs involved came from 1 country, some non-randomized trials or retrospective studies in other countries [9–14] had come up with similar conclusions. The usage of relatively aggressive drawing strategies other than 100 ml/day is recommended after looking back to those trails from other countries. As a result, the conclusion of this study may also be helpful for clinical workers of other countries. Besides that, this coincidence may lead to smaller heterogeneity in the surgical technique and genomic differences. We still need further high-quality, multicentre, randomized, controlled trials from other countries and regions. The postoperative complications that were analysed in this review only included reaccumulation of pleural fluid and atelectasis that was caused by the limited data from current studies. These may not be sufficient to evaluate the overall effects of 2 drainage strategies. Further studies should focus on more complications and items that contribute to the assessment of those 2 strategies. The accumulated pleural fluid after CT removal was mentioned in 4 trails [4–7]. References [6, 7] recorded this outcome from CT removal to 1 week after discharge, Ref. [5] recorded the outcome in 1 week after CT removal, and Ref. [4] had not mentioned the period of pleural fluid reaccumulation. The short follow-up and difference in observation time may influence the meta-analysis result. Further trials should observe this outcome for 1 month at least after CT removal to verify the correlation between CT management and pleural fluid reaccumulation [1]. CONCLUSIONS The meta-analysis result showed that increasing the threshold of daily drainage to 300 ml before removing the CT showed the same clinically meaningful and statistically improvement in amount of drainage in the first 24 h after operation, the patient numbers of pleural fluid reaccumulation, the patient numbers of atelectasis after operation and the number of patients requiring thoracentesis after removing the CT; however, using the volume threshold of 300 ml significantly decreases the duration of CT drainage and the hospital stay after operation compared with patients removing the CT when the daily drainage is <100 ml. Hence, patients using the daily threshold of 300 ml before removing the CT may be more suitable for patients after a pulmonary lobectomy. Although there is convincing evidence to confirm the results mentioned herein, they still need to be confirmed by large-sample, multicentre, randomized, controlled trials. SUPPLEMENTARY MATERIAL Supplementary material is available at ICVTS online. Funding This study was supported by Beijing Municipal Administration of Hospitals Ascent Plan [DFL20151501]. Conflict of interest: none declared. REFERENCES 1 Brunelli A , Beretta E , Cassivi SD , Cerfolio RJ , Detterbeck F , Kiefer T et al. Consensus definitions to promote an evidence-based approach to management of the pleural space. A collaborative proposal by ESTS, AATS, STS, and GTSC . Eur J Cardiothorac Surg 2011 ; 40 : 291 – 7 . Google Scholar CrossRef Search ADS PubMed 2 Gomez-Caro A , Roca MJ , Torres J , Cascales P , Terol E , Castaner J et al. Successful use of a single chest drain postlobectomy instead of two classical drains: a randomized study . Eur J Cardiothorac Surg 2006 ; 29 : 562 – 6 . Google Scholar CrossRef Search ADS PubMed 3 Wang J , Ni B , Ma H. The application of fast-track surgery in thoracoscopic lung cancer operation for the aged patients . J Laparosc Surg 2015 ; 8 : 581 – 5 . 4 Zhang W , Liu K , Pei Y , Zhao J. Improved chest tube application promotes fast track recovery after lobectomy . Chin J Front Med Sci (Elect Ver) 2015 ; 7 : 56 – 8 . 5 Han H , Zhang X , Wang D , Yao P. The association between drainage volume and removal of chest tube after video-assisted thoracoscopic lobectomy . Tianjin Med J 2015 ; 1 : 85 – 7 . 6 Li W , Jin Z , Duan DK. Comparison of the timing of thoracic drainage tube removal in gerontal patients after lung cancer surgery . Chin J Gerontol 2014 ; 16 : 4534 – 6 . 7 Zhang Y , Li H , Hu B , Li T , Miao J-B , You B et al. A prospective randomized single-blind control study of volume threshold for chest tube removal following lobectomy . World J Surg 2014 ; 38 : 60 – 7 . Google Scholar CrossRef Search ADS PubMed 8 Teng YJ , Pan SM , Liu YL , Yang KH , Zhang YC , Tian JH et al. A meta-analysis of randomized controlled trials of fixation versus nonfixation of mesh in laparoscopic total extraperitoneal inguinal hernia repair . Surg Endosc 2011 ; 25 : 2849 – 58 . Google Scholar CrossRef Search ADS PubMed 9 Bjerregaard LS , Jensen K , Petersen RH , Hansen HJ. Early chest tube removal after video-assisted thoracic surgery lobectomy with serous fluid production up to 500 ml/day . Eur J Cardiothorac Surg 2014 ; 45 : 241 – 6 . Google Scholar CrossRef Search ADS PubMed 10 Younes RN , Gross JL , Aguiar S , Haddad FJ , Deheinzelin D. When to remove a chest tube? A randomized study with subsequent prospective consecutive validation . J Am Coll Surg 2002 ; 195 : 658 – 62 . Google Scholar CrossRef Search ADS PubMed 11 Hessami MA , Najafi F , Hatami S. Volume threshold for chest tube removal: a randomized controlled trial . J Inj Violence Res 2009 ; 1 : 33 – 6 . Google Scholar CrossRef Search ADS PubMed 12 McKenna RJ , Mahtabifard A , Pickens A , Kusuanco D , Fuller CB. Fast-tracking after video-assisted thoracoscopic surgery lobectomy, segmentectomy, and pneumonectomy . Ann Thorac Surg 2007 ; 84 : 1663 – 7; discussion 1667–8. Google Scholar CrossRef Search ADS PubMed 13 Cerfolio RJ , Bryant AS. Results of a prospective algorithm to remove chest tubes after pulmonary resection with high output . J Thorac Cardiovasc Surg 2008 ; 135 : 269 – 73 . Google Scholar CrossRef Search ADS PubMed 14 Göttgens KW , Siebenga J , Belgers EH , van Huijstee PJ , Bollen EC. Early removal of the chest tube after complete video-assisted thoracoscopic lobectomies . Eur J Cardiothorac Surg 2011 ; 39 : 575 – 8 . Google Scholar CrossRef Search ADS PubMed 15 Xie HY , Xu K , Tang JX , Bian W , Ma HT , Zhao J et al. A prospective randomized, controlled trial deems a drainage of 300 ml/day safe before removal of the last chest drain after video-assisted thoracoscopic surgery lobectomy . Interact CardioVasc Thorac Surg 2015 ; 21 : 200 – 5 . Google Scholar CrossRef Search ADS PubMed © The Author(s) 2018. Published by Oxford University Press on behalf of the European Association for Cardio-Thoracic Surgery. All rights reserved. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices)

Journal

Interactive CardioVascular and Thoracic SurgeryOxford University Press

Published: May 7, 2018

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off