BEST EVIDENCE TOPIC – THORACIC
Interactive CardioVascular and Thoracic Surgery 21 (2015) 379–383 doi:10.1093/icvts/ivv149 Advance Access publication 11 June 2015
Cite this article as: Toufektzian L, Patris V, Sepsas E, Konstantinou M. Does postoperative mechanical ventilation predispose to bronchopleural fistula formation in patients undergoing pneumonectomy? Interact CardioVasc Thorac Surg 2015;21:379–83.
Does postoperative mechanical ventilation predispose to bronchopleural fistula formation in patients undergoing pneumonectomy? Levon Toufektziana,*, Vasileios Patrisb, Evangelos Sepsasc and Marios Konstantinouc a b c
Department of Thoracic Surgery, Guy’s Hospital, London, UK Department of Cardiothoracic Surgery, Liverpool Heart and Chest Hospital, Liverpool, UK Department of Thoracic Surgery, “Sotiria” General Hospital for Chest Diseases, Athens, Greece
* Corresponding author. Department of Thoracic Surgery, Guy’s Hospital, London, UK. Tel: +44-7902-613552; e-mail: [email protected] (L. Toufektzian). Received 16 August 2014; received in revised form 14 April 2015; accepted 18 May 2015
A best evidence topic was written according to a structured protocol. The question addressed was whether postoperative mechanical ventilation has any effect on the incidence of development of bronchopleural fistulas (BPFs) in patients undergoing pneumonectomy. A total of 40 papers were identified using the reported search, of which 8, all retrospective, represented the best evidence to answer the clinical question. The authors, date, journal, country, study type, population, outcomes and key results are tabulated. Of the eight identified papers, six of them reported a statistically significant relationship between postoperative mechanical ventilation and the occurrence of bronchopleural fistula in patients undergoing pneumonectomy (P = 0.027–0.0001). In two of these studies, postoperative mechanical ventilation was identified during multivariate analysis as an independent predictor for the development of BPF after pneumonectomy (odds ratio 15.57 and 33.1), indicating a causal relationship whereas, in the other four reports, statistical significance was the result of univariate analysis. In another study, the difference between these two groups approached but did not reach statistical significance (P = 0.057). Finally, one study reported no association between postoperative mechanical ventilation and the development of post-pneumonectomy BPF (0.16). Apart from mechanical ventilation, pre-existing pleuropulmonary infection was reported by one study as an independent predictor for the development of post-pneumonectomy BPF whereas, in two other studies, its impact approached but did not reach statistical significance. Another study did not find any association between preoperative infection and postoperative BPF occurrence. In conclusion, the majority of the reported studies report a significant relationship between mechanical ventilation after pneumonectomy and the occurrence of BPF. Every effort should be made to achieve extubation at the earliest possible time to withdraw the effects of the continuous barotrauma on the bronchial stump, although its impact cannot be quantified. Performing pneumonectomy in the presence of infectious conditions may contribute to the development of postoperative BPF, but its role is less well defined. Keywords: Bronchopleural fistula • Pneumonectomy • Mechanical ventilation • Risk factors
INTRODUCTION A best evidence topic was constructed according to a structured protocol. This is fully described in the ICVTS [1].
THREE-PART QUESTION In patients [undergoing pneumonectomy] does [the use of postoperative mechanical ventilation] increase the [risk for the development of bronchopleural fistula]?
CLINICAL SCENARIO A 59-year old lady is referred for consideration for left pneumonectomy due to a centrally located squamous cell carcinoma.
She underwent induction chemotherapy for a positive left paratracheal lymph node diagnosed with trans-bronchial needle aspiration, and repeat mediastinal staging with mediastinoscopy did not reveal any residual disease. Although, the tumour has shrunk as a result of the induction treatment, the need for pneumonectomy cannot be avoided. Her diffusing capacity for carbon monoxide has significantly reduced as a result of chemotherapy from 78 to 61% of predicted. Owing to her diminished respiratory reserve, concerns are raised that she is likely to need a period of postoperative mechanical ventilation, possibly increasing her risk for the development of bronchopleural fistula (BPF). If the risk of BPF is increased as a result of assisted mechanical ventilation in the postoperative period, bronchial stump reinforcement is strongly advised. You resolve to check the literature.
© The Author 2015. Published by Oxford University Press on behalf of the European Association for Cardio-Thoracic Surgery. All rights reserved.
BEST EVIDENCE TOPIC
Abstract
L. Toufektzian et al. / Interactive CardioVascular and Thoracic Surgery
380
Table 1: Best evidence papers Author, date, journal and country Study type (level of evidence)
Patient group
Outcomes
Key results
Comments
Hu et al. (2013), Ann Thorac Surg, China [2]
Study period: 1995–2010
Incidence of BPF
Group A: 23 (3.8%) Group B: 7 (9.3%) P = 0.027
Mechanical ventilation after pneumonectomy was found to be a risk factor for BPF
Study period: 1990–99
Incidence of BPF
n = 12 (6.8%)
175 patients undergoing pneumonectomy for NSCLC. 135 of them did not receive postoperative mechanical ventilation (Group A), whereas 40 patients did (Group B). 25 patients underwent pneumonectomy in the presence of pleuropulmonary infection (14.3%)
Incidence of BPF with regard to postoperative mechanical ventilation
Group A: 5/135 (3.7%) Group B: 7/40 (17.5%) P = 0.001
The incidence of BPF was significantly higher in patients submitted to postoperative mechanical ventilation
Incidence of BPF with regard to pleuropulmonary infection
2/25 (8%) P = 0.06
Study period: 1986–97
Incidence of BPF with regard to postoperative mechanical ventilation
Univariate analysis Group A: 1/96 (1%) Group B: 12/146 (8.2%) P = 0.015
Retrospective cohort study (level 2b) Sirbu et al. (2001), Ann Thorac Cardiovasc Surg, Germany [3] Retrospective cohort study (level 2b)
Algar et al. (2001), Ann Thorac Surg, Spain [4] Retrospective cohort study (level 2b)
de Perrot et al. (1999), Scand Cardiovasc J, Switzerland [5] Retrospective cohort study (level 2b)
Wright et al. (1996), J Thorac Cardiovasc Surg, USA [6] Retrospective cohort study (level 2b)
Birdas et al. (2012), Ann Surg Oncol, USA [7] Retrospective cohort study (level 2b)
684 patients undergoing pneumonectomy for NSCLC. 609 of them did not receive postoperative mechanical ventilation (Group A) and 75 did (Group B)
242 patients undergoing pneumonectomy for lung cancer. Early extubation was achieved in 96 of them (Group A, 39.7%), whereas 146 (Group B, 60.3%) cases were under mechanical ventilation for several hours after the operation
Multivariate analysis β = 2.7453 SE = 1.1974 Wald = 5.256 P = 0.0219 OR = 15.57 Incidence of BPF (with/without postoperative ventilation)
3/4
Impact of postoperative ventilation on the incidence of post-pneumonectomy BPF
Univariate analysis P < 0.01
Incidence of BPF with regard to the presence of preoperative infection
3/7 (43%) P > 0.05
Study period: 1980–95
Incidence of BPF
n = 8 (3.1%)
256 patients undergoing pneumonectomy for malignant and benign aetiologies. 225 patients did not receive postoperative mechanical ventilation (Group A), whereas 31 patients did (Group B). 37 patients presented with pleuropulmonary infection (14.5%)
Incidence of BPF with regard to postoperative mechanical ventilation
Group A: 2/225 (0.9%) Group B: 6/31 (19.3%) P = 0.0001
Incidence of BPF with regard to pleuropulmonary infection
3/37 (8.1%) P = 0.06
Study period: 1992–2010
Incidence of BPF
n = 11 (7.6%)
145 patients undergoing pneumonectomy for malignant and benign aetiologies. 53 of them received postoperative mechanical ventilation
Impact of length of postoperative ventilation on the incidence of post-pneumonectomy BPF
Multivariate analysis P = 0.057 OR = 1.04 95% CI: 0.99–1.10
Study period: 1990–96 100 patients undergoing pneumonectomy for lung cancer. 7 cases performed in the presence of preoperative infection
Multivariate analysis OR = 33.1 95% CI: 6.9–157.6
Pleuropulmonary infection approached statistical significance as a contributing factor for the development of postpneumonectomy BPF Postoperative mechanical ventilation was found to be an independent predictor of BPF in the multivariate analysis. Based on these results, the early extubation of these patients may be advisable to prevent early BPF onset
The need for postoperative ventilation was an independent risk factor for the development of BPF. However, it is likely that other risk factors, such as systematic mediastinal lymph node dissection and bronchial stapling were involved
If one anticipates the need for postoperative ventilation in a patient who has undergone pneumonectomy, special attention should be paid to the airway closure and to buttressing The presence of pleuropulmonary infection might constitute a contributing factor although it did not reach statistical significance Prolonged mechanical ventilation after pneumonectomy was a variable that approached statistical significance as an independent risk factor for BPF
Continued
L. Toufektzian et al. / Interactive CardioVascular and Thoracic Surgery
381
Table 1: (Continued) Author, date, journal and country Study type (level of evidence)
Patient group
Outcomes
Key results
Comments
Panagopoulos et al. (2009), Interact CardioVasc Thorac Surg, Greece [8]
Study period: 1999–2005
Incidence of BPF
n = 5 (2%)
221 patients undergoing pneumonectomy for NSCLC. 13 patients presented with respiratory infection (5.9%)
Development of respiratory failure requiring mechanical ventilation in the ICU in association with the occurrence of BPF
no BPF: 11/216 (5%) BPF: 2/5 (40%) P = 0.029
Apart from postoperative mechanical ventilation, the development of BPF was significantly correlated with the presence of preoperative pleuropulmonary infection
Incidence of respiratory infection in association with the occurrence of BPF
no BPF: 9/216 (4.2%) BPF: 4/5 (80%) P < 0.001
Usage of prolonged postoperative ventilation with regard to the development of postoperative BPF
Group A (BPF): 1/5 Group B (no BPF): 6/194 P = 0.16
Retrospective cohort study (level 2b)
Hubaut et al. (1999), Eur J Cardiothorac Surg, France [9] Retrospective cohort study (level 2b)
Study period: 1988–97 199 patients undergoing pneumonectomy by the same operator. 5 patients developed postoperative BPF (Group A, 2.5%), whereas 194 patients did not (Group B, 97.5%)
No association was found between the occurrence of post-pneumonectomy bronchopleural fistula and prolonged assisted ventilation (>24 h)
SEARCH STRATEGY Medline (R) from 1946 to August 2014 using the Ovid interface. Search strategy employed: (exp Pneumonectomy/) AND [(exp Bronchial Fistula/) odds ratio (OR) (bronchopleural fistula.mp.)] AND [(exp Respiration, Artificial/) OR (mechanical ventilation.mp.)].
SEARCH OUTCOMES The reported search yielded 40 results. All relevant papers were screened and their reference lists were cross-checked. This process extracted eight papers that were deemed to offer the best evidence. The papers are detailed in Table 1.
RESULTS Hu et al. [2] retrospectively evaluated 684 patients undergoing pneumonectomy for non-small-cell lung cancer (NSCLC) attempting to generate a clinical risk model for the occurrence of BPF. Six hundred and nine patients did not have postoperative mechanical ventilation (Group A), whereas 75 did (Group B). The incidence of BPF was significantly higher in those patients receiving postoperative mechanical ventilation (Group A: 23, 3.8% vs Group B: 7, 9.3%, P = 0.027). Sirbu et al. [3] retrospectively reviewed 175 patients who underwent pneumonectomy for NSCLC. One hundred and thirty-five of them did not receive postoperative mechanical ventilation (Group A), whereas 40 of them did (Group B). The incidence of BPF was significantly higher in patients who had postoperative assisted ventilation (Group A: 5, 3.7% vs Group B: 7, 17.5%, P = 0.001). Furthermore, 25 of 175 patients undergoing pneumonectomy presented with
preoperative pleuropulmonary infection. The incidence of postoperative BPF in this group of patients was 8% (2/25, P = 0.06). Algar et al. [4] performed a retrospective analysis of the risk factors for the development of BPF in 242 patients undergoing pneumonectomy for lung cancer. In this cohort, early extubation was achieved in 96 of them (Group A), whereas 146 patients required postoperative mechanical ventilation for at least a few hours after the completion of the procedure (Group B). The occurrence of BPF was significantly more likely for those patients receiving postoperative mechanical ventilation (Group A: 1, 1% vs Group B: 12, 8.2%, P = 0.015). Multivariate analysis identified postoperative ventilation as a risk factor for the development of BPF in patients undergoing pneumonectomy (β = 2.7453, OR = 15.57, P = 0.0219). De Perrot et al. [5] reviewed the outcomes from 100 consecutive patients undergoing pneumonectomy for lung cancer. Univariate (P < 0.01) and multivariate analysis [OR = 33.1, 95% confidence interval (CI): 6.9–157.6] indicated that postoperative mechanical ventilation had a significant impact on the occurrence of postpneumonectomy BPF. According to the authors of the study, performing a pneumonectomy under infectious conditions did not seem to affect the risk of postoperative BPF (P > 0.05). Wright et al. [6] retrospectively evaluated 256 patients undergoing pneumonectomy for malignant and benign aetiologies. Early extubation immediately after the conclusion of the procedure was achieved in 225 patients (Group A), whereas in 31 patients a period of postoperative mechanical ventilation was considered necessary (Group B). The incidence of BPF was significantly increased in those patients requiring assisted ventilation (Group A: 2, 0.9% vs Group B: 6, 19.3%, P = 0.0001). In addition, 37 of 256 patients undergoing pneumonectomy presented with preoperative pleuropulmonary infection. The incidence of postoperative BPF in this group of patients was 8.1% (3/37, P = 0.06).
BEST EVIDENCE TOPIC
NSCLC: non-small-cell lung cancer; BPFs: bronchopleural fistulas; OR: odds ratio; CI: confidence interval; SE: standard error.
382
L. Toufektzian et al. / Interactive CardioVascular and Thoracic Surgery
Birdas et al. [7] analysed retrospectively the results from 145 patients undergoing pneumonectomy for malignant and benign aetiologies, 53 of whom received postoperative mechanical ventilation. The authors concluded that prolonged mechanical ventilation approached statistical significance as an independent risk factor for BPF (OR = 1.04, 95% CI: 0.99–1.10, P = 0.057). In a retrospective review of 221 patients undergoing pneumonectomy, Panagopoulos et al. [8] reported that mechanical ventilation was employed in significantly more patients who developed (n = 2, 40%) compared with those who did not develop BPF (n = 11, 5%) (P = 0.029). Of 5 patients in total who developed postoperative BPF, 4 underwent pneumonectomy in the presence of respiratory infection (80%), as opposed to 9/216 patients with preoperative respiratory infection who underwent pneumonectomy and did not develop postoperative BPF (4.2%) (P < 0.0001). Finally, Hubaut et al. [9] performed a retrospective study on 209 patients undergoing pneumonectomy. There was an intraoperative death, whereas in 9 patients data were missing. Of the 199 remaining patients, 5 developed a postoperative BPF. In the group of patients who developed a postoperative BPF, only 1 had had prolonged postoperative mechanical ventilation (1/5, 20%), when compared with 6 patients who required mechanical ventilation in the group of those who did not develop a postoperative BPF (6/ 194, 3.1%) (P = 0.16).
CLINICAL BOTTOM LINE In conclusion, the majority of the reported studies report a significant relationship between mechanical ventilation after pneumonectomy and the occurrence of BPF. Two of the eight presented studies report a causal relationship, whereas from the remaining six studies, four report a significant correlation between the use of postoperative mechanical ventilation and the development of post-pneumonectomy BPF. Every effort should be made to achieve extubation at the earliest possible time to withdraw the effects of the continuous barotrauma on the bronchial stump, although its impact cannot be quantified. Performing a pneumonectomy in the presence of infectious conditions may contribute to the development of postoperative BPF; however, its role is less well defined. Conflict of interest: none declared.
[6] Wright CD, Wain JC, Mathisen DJ, Grillo HC. Postpneumonectomy bronchopleural fistula after sutured bronchial closure: incidence, risk factors, and management. J Thorac Cardiovasc Surg 1996;112:1367–71. [7] Birdas TJ, Morad MH, Okereke IC, Rieger KM, Kruter LE, Mathur PN et al. Risk factors for bronchopleural fistula after right pneumonectomy: does eliminating the stump diverticulum provide protection? Ann Surg Oncol 2012;19:1336–42. [8] Panagopoulos ND, Apostolakis E, Koletsis E, Prokakis C, Hountis P, Sakellaropoulos G et al. Low incidence of bronchopleural fistula after pneumonectomy for lung cancer. Interact CardioVasc Thorac Surg 2009;9: 571–5. [9] Hubaut JJ, Baron O, Al Habash O, Despins P, Duveau D, Michaud JL. Closure of the bronchial stump by manual suture and incidence of bronchopleural fistula in a series of 209 pneumonectomies for lung cancer. Eur J Cardiothorac Surg 1999;16:418–23.
eComment. Is the problem the duration of mechanical ventilation or how it is performed? Authors: Marco Chiappetta, Dania Nachira, Venanzio Porziella and Stefano Margaritora Department of General Thoracic Surgery, Catholic University, Rome, Italy doi: 10.1093/icvts/ivv211 © The Author 2015. Published by Oxford University Press on behalf of the European Association for Cardio-Thoracic Surgery. All rights reserved. We read with interest the article of Toufektzian et al. [1] on the development of bronchopleural fistula after prolonged mechanical ventilation. In our experience we also have found a higher incidence of bronchopleural fistula in pneumonectomies after prolonged mechanical ventilation (P = 0.001), but we had defined “prolonged ventilation” as one that was longer than 24 h and when attempts to remove the orotracheal tube had failed. We think that in the literature, there is a high variability of this definition. For example, Wright et al. performed early extubation at the end of surgery [2], but in our experience, we noted that is not always possible, because one has to consider the timing of the procedure and the patient’s condition. Another point of interest is that the ventilation protocol is not the same for every patient. In fact, many variables could affect the ventilation: volume support, pressure support, positive end-expiratory pressure, respiratory rate etc. These are often chosen by the anaesthesiologist during the surgery, and this could vary from centre to centre. In fact, the incidence of bronchopleural fistula is not 100% in patients undergoing prolonged mechanical ventilation, but fistulas occur only in a small subset of patients. Thus, the duration of the ventilation is probably not the only risk factor during invasive ventilation [3,4], but is variable regardless of the length of ventilation [3,4]. Perhaps the right question should be: is the problem the duration of mechanical ventilation or how it is performed? Finally, we also noted a higher incidence of fistulas in patients with preoperative lung infection, with no statistical significance at multivariate analysis, as other authors have observed [4,5]. In this case, we think that it is a correlate problem, because preoperative lung infection could create parenchymal damage and aggravate respiratory functionality, increasing the risk of prolonged mechanical ventilation. Based on the data reported, we would really appreciate the Authors’ reflections and reaction. Conflict of interest: none declared.
REFERENCES [1] Dunning J, Prendergast B, Mackway-Jones K. Towards evidence-based medicine in cardiothoracic surgery: best BETS. Interact CardioVasc Thorac Surg 2003;2:405–9. [2] Hu XF, Duan L, Jiang GN, Wang H, Liu HC, Chen C. A clinical risk model for the evaluation of bronchopleural fistula in non-small cell lung cancer after pneumonectomy. Ann Thorac Surg 2013;96:419–24. [3] Sirbu H, Busch T, Aleksic I, Schreiner W, Oster O, Dalichau H. Bronchopleural fistula in the surgery of non-small cell lung cancer: incidence, risk factors, and management. Ann Thorac Cardiovasc Surg 2001;7: 330–6. [4] Algar FJ, Alvarez A, Aranda JL, Salvatierra A, Baamonde C, López-Pujol FJ. Prediction of early bronchopleural fistula after pneumonectomy: a multivariate analysis. Ann Thorac Surg 2001;72:1662–7. [5] de Perrot M, Licker M, Robert J, Spiliopoulos A. Incidence, risk factors and management of bronchopleural fistulae after pneumonectomy. Scand Cardiovasc J 1999;33:171–4.
References [1] Toufektzian L, Patris V, Sepsas E, Konstantinou M. Does postoperative mechanical ventilation predispose to bronchopleural fistula formation in patients undergoing pneumonectomy? Interact CardioVasc Thorac Surg 2015;21:379–83. [2] Wright CD, Wain JC, Mathisen DJ, Grillo HC. Postpneumonectomy bronchopleural fistula after sutured bronchial closure: incidence, risk factors, and management. J Thorac Cardiovasc Surg 1996;112:1367–71. [3] Algar FJ, Alvarez A, Aranda JL, Salvatierra A, Baamonde C, Lopez-Pujol FJ. Prediction of early bronchopleural fistula after pneumonectomy: a multivariate analysis. Ann Thorac Surg 2001;72:1662–7. [4] de Perrot M, Licker M, Robert J, Spiliopoulos A. Incidence, risk factors and management of bronchopleural fistulae after pneumonectomy. Scand Cardiovasc J 1999;33:171–4. [5] Sirbu H, Busch T, Aleksic I, Schreiner W, Oster O, Dalichau H. Bronchopleural fistula in the surgery of non-small cell lung cancer: incidence, risk factors, and management. Ann Thorac Cardiovasc Surg 2001;7: 330–6.
Interactive CardioVascular and Thoracic Surgery 21 (2015) 379–383 doi:10.1093/icvts/ivv149 Advance Access publication 11 June 2015
Cite this article as: Toufektzian L, Patris V, Sepsas E, Konstantinou M. Does postoperative mechanical ventilation predispose to bronchopleural fistula formation in patients undergoing pneumonectomy? Interact CardioVasc Thorac Surg 2015;21:379–83.
Does postoperative mechanical ventilation predispose to bronchopleural fistula formation in patients undergoing pneumonectomy? Levon Toufektziana,*, Vasileios Patrisb, Evangelos Sepsasc and Marios Konstantinouc a b c
Department of Thoracic Surgery, Guy’s Hospital, London, UK Department of Cardiothoracic Surgery, Liverpool Heart and Chest Hospital, Liverpool, UK Department of Thoracic Surgery, “Sotiria” General Hospital for Chest Diseases, Athens, Greece
* Corresponding author. Department of Thoracic Surgery, Guy’s Hospital, London, UK. Tel: +44-7902-613552; e-mail: [email protected] (L. Toufektzian). Received 16 August 2014; received in revised form 14 April 2015; accepted 18 May 2015
A best evidence topic was written according to a structured protocol. The question addressed was whether postoperative mechanical ventilation has any effect on the incidence of development of bronchopleural fistulas (BPFs) in patients undergoing pneumonectomy. A total of 40 papers were identified using the reported search, of which 8, all retrospective, represented the best evidence to answer the clinical question. The authors, date, journal, country, study type, population, outcomes and key results are tabulated. Of the eight identified papers, six of them reported a statistically significant relationship between postoperative mechanical ventilation and the occurrence of bronchopleural fistula in patients undergoing pneumonectomy (P = 0.027–0.0001). In two of these studies, postoperative mechanical ventilation was identified during multivariate analysis as an independent predictor for the development of BPF after pneumonectomy (odds ratio 15.57 and 33.1), indicating a causal relationship whereas, in the other four reports, statistical significance was the result of univariate analysis. In another study, the difference between these two groups approached but did not reach statistical significance (P = 0.057). Finally, one study reported no association between postoperative mechanical ventilation and the development of post-pneumonectomy BPF (0.16). Apart from mechanical ventilation, pre-existing pleuropulmonary infection was reported by one study as an independent predictor for the development of post-pneumonectomy BPF whereas, in two other studies, its impact approached but did not reach statistical significance. Another study did not find any association between preoperative infection and postoperative BPF occurrence. In conclusion, the majority of the reported studies report a significant relationship between mechanical ventilation after pneumonectomy and the occurrence of BPF. Every effort should be made to achieve extubation at the earliest possible time to withdraw the effects of the continuous barotrauma on the bronchial stump, although its impact cannot be quantified. Performing pneumonectomy in the presence of infectious conditions may contribute to the development of postoperative BPF, but its role is less well defined. Keywords: Bronchopleural fistula • Pneumonectomy • Mechanical ventilation • Risk factors
INTRODUCTION A best evidence topic was constructed according to a structured protocol. This is fully described in the ICVTS [1].
THREE-PART QUESTION In patients [undergoing pneumonectomy] does [the use of postoperative mechanical ventilation] increase the [risk for the development of bronchopleural fistula]?
CLINICAL SCENARIO A 59-year old lady is referred for consideration for left pneumonectomy due to a centrally located squamous cell carcinoma.
She underwent induction chemotherapy for a positive left paratracheal lymph node diagnosed with trans-bronchial needle aspiration, and repeat mediastinal staging with mediastinoscopy did not reveal any residual disease. Although, the tumour has shrunk as a result of the induction treatment, the need for pneumonectomy cannot be avoided. Her diffusing capacity for carbon monoxide has significantly reduced as a result of chemotherapy from 78 to 61% of predicted. Owing to her diminished respiratory reserve, concerns are raised that she is likely to need a period of postoperative mechanical ventilation, possibly increasing her risk for the development of bronchopleural fistula (BPF). If the risk of BPF is increased as a result of assisted mechanical ventilation in the postoperative period, bronchial stump reinforcement is strongly advised. You resolve to check the literature.
© The Author 2015. Published by Oxford University Press on behalf of the European Association for Cardio-Thoracic Surgery. All rights reserved.
BEST EVIDENCE TOPIC
Abstract
L. Toufektzian et al. / Interactive CardioVascular and Thoracic Surgery
380
Table 1: Best evidence papers Author, date, journal and country Study type (level of evidence)
Patient group
Outcomes
Key results
Comments
Hu et al. (2013), Ann Thorac Surg, China [2]
Study period: 1995–2010
Incidence of BPF
Group A: 23 (3.8%) Group B: 7 (9.3%) P = 0.027
Mechanical ventilation after pneumonectomy was found to be a risk factor for BPF
Study period: 1990–99
Incidence of BPF
n = 12 (6.8%)
175 patients undergoing pneumonectomy for NSCLC. 135 of them did not receive postoperative mechanical ventilation (Group A), whereas 40 patients did (Group B). 25 patients underwent pneumonectomy in the presence of pleuropulmonary infection (14.3%)
Incidence of BPF with regard to postoperative mechanical ventilation
Group A: 5/135 (3.7%) Group B: 7/40 (17.5%) P = 0.001
The incidence of BPF was significantly higher in patients submitted to postoperative mechanical ventilation
Incidence of BPF with regard to pleuropulmonary infection
2/25 (8%) P = 0.06
Study period: 1986–97
Incidence of BPF with regard to postoperative mechanical ventilation
Univariate analysis Group A: 1/96 (1%) Group B: 12/146 (8.2%) P = 0.015
Retrospective cohort study (level 2b) Sirbu et al. (2001), Ann Thorac Cardiovasc Surg, Germany [3] Retrospective cohort study (level 2b)
Algar et al. (2001), Ann Thorac Surg, Spain [4] Retrospective cohort study (level 2b)
de Perrot et al. (1999), Scand Cardiovasc J, Switzerland [5] Retrospective cohort study (level 2b)
Wright et al. (1996), J Thorac Cardiovasc Surg, USA [6] Retrospective cohort study (level 2b)
Birdas et al. (2012), Ann Surg Oncol, USA [7] Retrospective cohort study (level 2b)
684 patients undergoing pneumonectomy for NSCLC. 609 of them did not receive postoperative mechanical ventilation (Group A) and 75 did (Group B)
242 patients undergoing pneumonectomy for lung cancer. Early extubation was achieved in 96 of them (Group A, 39.7%), whereas 146 (Group B, 60.3%) cases were under mechanical ventilation for several hours after the operation
Multivariate analysis β = 2.7453 SE = 1.1974 Wald = 5.256 P = 0.0219 OR = 15.57 Incidence of BPF (with/without postoperative ventilation)
3/4
Impact of postoperative ventilation on the incidence of post-pneumonectomy BPF
Univariate analysis P < 0.01
Incidence of BPF with regard to the presence of preoperative infection
3/7 (43%) P > 0.05
Study period: 1980–95
Incidence of BPF
n = 8 (3.1%)
256 patients undergoing pneumonectomy for malignant and benign aetiologies. 225 patients did not receive postoperative mechanical ventilation (Group A), whereas 31 patients did (Group B). 37 patients presented with pleuropulmonary infection (14.5%)
Incidence of BPF with regard to postoperative mechanical ventilation
Group A: 2/225 (0.9%) Group B: 6/31 (19.3%) P = 0.0001
Incidence of BPF with regard to pleuropulmonary infection
3/37 (8.1%) P = 0.06
Study period: 1992–2010
Incidence of BPF
n = 11 (7.6%)
145 patients undergoing pneumonectomy for malignant and benign aetiologies. 53 of them received postoperative mechanical ventilation
Impact of length of postoperative ventilation on the incidence of post-pneumonectomy BPF
Multivariate analysis P = 0.057 OR = 1.04 95% CI: 0.99–1.10
Study period: 1990–96 100 patients undergoing pneumonectomy for lung cancer. 7 cases performed in the presence of preoperative infection
Multivariate analysis OR = 33.1 95% CI: 6.9–157.6
Pleuropulmonary infection approached statistical significance as a contributing factor for the development of postpneumonectomy BPF Postoperative mechanical ventilation was found to be an independent predictor of BPF in the multivariate analysis. Based on these results, the early extubation of these patients may be advisable to prevent early BPF onset
The need for postoperative ventilation was an independent risk factor for the development of BPF. However, it is likely that other risk factors, such as systematic mediastinal lymph node dissection and bronchial stapling were involved
If one anticipates the need for postoperative ventilation in a patient who has undergone pneumonectomy, special attention should be paid to the airway closure and to buttressing The presence of pleuropulmonary infection might constitute a contributing factor although it did not reach statistical significance Prolonged mechanical ventilation after pneumonectomy was a variable that approached statistical significance as an independent risk factor for BPF
Continued
L. Toufektzian et al. / Interactive CardioVascular and Thoracic Surgery
381
Table 1: (Continued) Author, date, journal and country Study type (level of evidence)
Patient group
Outcomes
Key results
Comments
Panagopoulos et al. (2009), Interact CardioVasc Thorac Surg, Greece [8]
Study period: 1999–2005
Incidence of BPF
n = 5 (2%)
221 patients undergoing pneumonectomy for NSCLC. 13 patients presented with respiratory infection (5.9%)
Development of respiratory failure requiring mechanical ventilation in the ICU in association with the occurrence of BPF
no BPF: 11/216 (5%) BPF: 2/5 (40%) P = 0.029
Apart from postoperative mechanical ventilation, the development of BPF was significantly correlated with the presence of preoperative pleuropulmonary infection
Incidence of respiratory infection in association with the occurrence of BPF
no BPF: 9/216 (4.2%) BPF: 4/5 (80%) P < 0.001
Usage of prolonged postoperative ventilation with regard to the development of postoperative BPF
Group A (BPF): 1/5 Group B (no BPF): 6/194 P = 0.16
Retrospective cohort study (level 2b)
Hubaut et al. (1999), Eur J Cardiothorac Surg, France [9] Retrospective cohort study (level 2b)
Study period: 1988–97 199 patients undergoing pneumonectomy by the same operator. 5 patients developed postoperative BPF (Group A, 2.5%), whereas 194 patients did not (Group B, 97.5%)
No association was found between the occurrence of post-pneumonectomy bronchopleural fistula and prolonged assisted ventilation (>24 h)
SEARCH STRATEGY Medline (R) from 1946 to August 2014 using the Ovid interface. Search strategy employed: (exp Pneumonectomy/) AND [(exp Bronchial Fistula/) odds ratio (OR) (bronchopleural fistula.mp.)] AND [(exp Respiration, Artificial/) OR (mechanical ventilation.mp.)].
SEARCH OUTCOMES The reported search yielded 40 results. All relevant papers were screened and their reference lists were cross-checked. This process extracted eight papers that were deemed to offer the best evidence. The papers are detailed in Table 1.
RESULTS Hu et al. [2] retrospectively evaluated 684 patients undergoing pneumonectomy for non-small-cell lung cancer (NSCLC) attempting to generate a clinical risk model for the occurrence of BPF. Six hundred and nine patients did not have postoperative mechanical ventilation (Group A), whereas 75 did (Group B). The incidence of BPF was significantly higher in those patients receiving postoperative mechanical ventilation (Group A: 23, 3.8% vs Group B: 7, 9.3%, P = 0.027). Sirbu et al. [3] retrospectively reviewed 175 patients who underwent pneumonectomy for NSCLC. One hundred and thirty-five of them did not receive postoperative mechanical ventilation (Group A), whereas 40 of them did (Group B). The incidence of BPF was significantly higher in patients who had postoperative assisted ventilation (Group A: 5, 3.7% vs Group B: 7, 17.5%, P = 0.001). Furthermore, 25 of 175 patients undergoing pneumonectomy presented with
preoperative pleuropulmonary infection. The incidence of postoperative BPF in this group of patients was 8% (2/25, P = 0.06). Algar et al. [4] performed a retrospective analysis of the risk factors for the development of BPF in 242 patients undergoing pneumonectomy for lung cancer. In this cohort, early extubation was achieved in 96 of them (Group A), whereas 146 patients required postoperative mechanical ventilation for at least a few hours after the completion of the procedure (Group B). The occurrence of BPF was significantly more likely for those patients receiving postoperative mechanical ventilation (Group A: 1, 1% vs Group B: 12, 8.2%, P = 0.015). Multivariate analysis identified postoperative ventilation as a risk factor for the development of BPF in patients undergoing pneumonectomy (β = 2.7453, OR = 15.57, P = 0.0219). De Perrot et al. [5] reviewed the outcomes from 100 consecutive patients undergoing pneumonectomy for lung cancer. Univariate (P < 0.01) and multivariate analysis [OR = 33.1, 95% confidence interval (CI): 6.9–157.6] indicated that postoperative mechanical ventilation had a significant impact on the occurrence of postpneumonectomy BPF. According to the authors of the study, performing a pneumonectomy under infectious conditions did not seem to affect the risk of postoperative BPF (P > 0.05). Wright et al. [6] retrospectively evaluated 256 patients undergoing pneumonectomy for malignant and benign aetiologies. Early extubation immediately after the conclusion of the procedure was achieved in 225 patients (Group A), whereas in 31 patients a period of postoperative mechanical ventilation was considered necessary (Group B). The incidence of BPF was significantly increased in those patients requiring assisted ventilation (Group A: 2, 0.9% vs Group B: 6, 19.3%, P = 0.0001). In addition, 37 of 256 patients undergoing pneumonectomy presented with preoperative pleuropulmonary infection. The incidence of postoperative BPF in this group of patients was 8.1% (3/37, P = 0.06).
BEST EVIDENCE TOPIC
NSCLC: non-small-cell lung cancer; BPFs: bronchopleural fistulas; OR: odds ratio; CI: confidence interval; SE: standard error.
382
L. Toufektzian et al. / Interactive CardioVascular and Thoracic Surgery
Birdas et al. [7] analysed retrospectively the results from 145 patients undergoing pneumonectomy for malignant and benign aetiologies, 53 of whom received postoperative mechanical ventilation. The authors concluded that prolonged mechanical ventilation approached statistical significance as an independent risk factor for BPF (OR = 1.04, 95% CI: 0.99–1.10, P = 0.057). In a retrospective review of 221 patients undergoing pneumonectomy, Panagopoulos et al. [8] reported that mechanical ventilation was employed in significantly more patients who developed (n = 2, 40%) compared with those who did not develop BPF (n = 11, 5%) (P = 0.029). Of 5 patients in total who developed postoperative BPF, 4 underwent pneumonectomy in the presence of respiratory infection (80%), as opposed to 9/216 patients with preoperative respiratory infection who underwent pneumonectomy and did not develop postoperative BPF (4.2%) (P < 0.0001). Finally, Hubaut et al. [9] performed a retrospective study on 209 patients undergoing pneumonectomy. There was an intraoperative death, whereas in 9 patients data were missing. Of the 199 remaining patients, 5 developed a postoperative BPF. In the group of patients who developed a postoperative BPF, only 1 had had prolonged postoperative mechanical ventilation (1/5, 20%), when compared with 6 patients who required mechanical ventilation in the group of those who did not develop a postoperative BPF (6/ 194, 3.1%) (P = 0.16).
CLINICAL BOTTOM LINE In conclusion, the majority of the reported studies report a significant relationship between mechanical ventilation after pneumonectomy and the occurrence of BPF. Two of the eight presented studies report a causal relationship, whereas from the remaining six studies, four report a significant correlation between the use of postoperative mechanical ventilation and the development of post-pneumonectomy BPF. Every effort should be made to achieve extubation at the earliest possible time to withdraw the effects of the continuous barotrauma on the bronchial stump, although its impact cannot be quantified. Performing a pneumonectomy in the presence of infectious conditions may contribute to the development of postoperative BPF; however, its role is less well defined. Conflict of interest: none declared.
[6] Wright CD, Wain JC, Mathisen DJ, Grillo HC. Postpneumonectomy bronchopleural fistula after sutured bronchial closure: incidence, risk factors, and management. J Thorac Cardiovasc Surg 1996;112:1367–71. [7] Birdas TJ, Morad MH, Okereke IC, Rieger KM, Kruter LE, Mathur PN et al. Risk factors for bronchopleural fistula after right pneumonectomy: does eliminating the stump diverticulum provide protection? Ann Surg Oncol 2012;19:1336–42. [8] Panagopoulos ND, Apostolakis E, Koletsis E, Prokakis C, Hountis P, Sakellaropoulos G et al. Low incidence of bronchopleural fistula after pneumonectomy for lung cancer. Interact CardioVasc Thorac Surg 2009;9: 571–5. [9] Hubaut JJ, Baron O, Al Habash O, Despins P, Duveau D, Michaud JL. Closure of the bronchial stump by manual suture and incidence of bronchopleural fistula in a series of 209 pneumonectomies for lung cancer. Eur J Cardiothorac Surg 1999;16:418–23.
eComment. Is the problem the duration of mechanical ventilation or how it is performed? Authors: Marco Chiappetta, Dania Nachira, Venanzio Porziella and Stefano Margaritora Department of General Thoracic Surgery, Catholic University, Rome, Italy doi: 10.1093/icvts/ivv211 © The Author 2015. Published by Oxford University Press on behalf of the European Association for Cardio-Thoracic Surgery. All rights reserved. We read with interest the article of Toufektzian et al. [1] on the development of bronchopleural fistula after prolonged mechanical ventilation. In our experience we also have found a higher incidence of bronchopleural fistula in pneumonectomies after prolonged mechanical ventilation (P = 0.001), but we had defined “prolonged ventilation” as one that was longer than 24 h and when attempts to remove the orotracheal tube had failed. We think that in the literature, there is a high variability of this definition. For example, Wright et al. performed early extubation at the end of surgery [2], but in our experience, we noted that is not always possible, because one has to consider the timing of the procedure and the patient’s condition. Another point of interest is that the ventilation protocol is not the same for every patient. In fact, many variables could affect the ventilation: volume support, pressure support, positive end-expiratory pressure, respiratory rate etc. These are often chosen by the anaesthesiologist during the surgery, and this could vary from centre to centre. In fact, the incidence of bronchopleural fistula is not 100% in patients undergoing prolonged mechanical ventilation, but fistulas occur only in a small subset of patients. Thus, the duration of the ventilation is probably not the only risk factor during invasive ventilation [3,4], but is variable regardless of the length of ventilation [3,4]. Perhaps the right question should be: is the problem the duration of mechanical ventilation or how it is performed? Finally, we also noted a higher incidence of fistulas in patients with preoperative lung infection, with no statistical significance at multivariate analysis, as other authors have observed [4,5]. In this case, we think that it is a correlate problem, because preoperative lung infection could create parenchymal damage and aggravate respiratory functionality, increasing the risk of prolonged mechanical ventilation. Based on the data reported, we would really appreciate the Authors’ reflections and reaction. Conflict of interest: none declared.
REFERENCES [1] Dunning J, Prendergast B, Mackway-Jones K. Towards evidence-based medicine in cardiothoracic surgery: best BETS. Interact CardioVasc Thorac Surg 2003;2:405–9. [2] Hu XF, Duan L, Jiang GN, Wang H, Liu HC, Chen C. A clinical risk model for the evaluation of bronchopleural fistula in non-small cell lung cancer after pneumonectomy. Ann Thorac Surg 2013;96:419–24. [3] Sirbu H, Busch T, Aleksic I, Schreiner W, Oster O, Dalichau H. Bronchopleural fistula in the surgery of non-small cell lung cancer: incidence, risk factors, and management. Ann Thorac Cardiovasc Surg 2001;7: 330–6. [4] Algar FJ, Alvarez A, Aranda JL, Salvatierra A, Baamonde C, López-Pujol FJ. Prediction of early bronchopleural fistula after pneumonectomy: a multivariate analysis. Ann Thorac Surg 2001;72:1662–7. [5] de Perrot M, Licker M, Robert J, Spiliopoulos A. Incidence, risk factors and management of bronchopleural fistulae after pneumonectomy. Scand Cardiovasc J 1999;33:171–4.
References [1] Toufektzian L, Patris V, Sepsas E, Konstantinou M. Does postoperative mechanical ventilation predispose to bronchopleural fistula formation in patients undergoing pneumonectomy? Interact CardioVasc Thorac Surg 2015;21:379–83. [2] Wright CD, Wain JC, Mathisen DJ, Grillo HC. Postpneumonectomy bronchopleural fistula after sutured bronchial closure: incidence, risk factors, and management. J Thorac Cardiovasc Surg 1996;112:1367–71. [3] Algar FJ, Alvarez A, Aranda JL, Salvatierra A, Baamonde C, Lopez-Pujol FJ. Prediction of early bronchopleural fistula after pneumonectomy: a multivariate analysis. Ann Thorac Surg 2001;72:1662–7. [4] de Perrot M, Licker M, Robert J, Spiliopoulos A. Incidence, risk factors and management of bronchopleural fistulae after pneumonectomy. Scand Cardiovasc J 1999;33:171–4. [5] Sirbu H, Busch T, Aleksic I, Schreiner W, Oster O, Dalichau H. Bronchopleural fistula in the surgery of non-small cell lung cancer: incidence, risk factors, and management. Ann Thorac Cardiovasc Surg 2001;7: 330–6.
