Documents

Reseccion Pulmonar en Hemoptisis Severa

Description
pdf
Categories
Published
of 10
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Related Documents
Share
Transcript
  ORIGINAL ARTICLES: GENERAL THORACICGENERAL THORACIC SURGERY: The Annals of Thoracic Surgery  CME Program is located online at http://cme.ctsnetjournals.org.To take the CME activity related to this article, you must have either an STS member or anindividual non-member subscription to the journal. Surgical Lung Resection for Severe Hemoptysis Claire Andréjak,  MD,  Antoine Parrot,  MD,  Bernard Bazelly,  MD, Pierre Yves Ancel,  MD,  Michel Djibré,  MD,  Antoine Khalil,  MD, Dominique Grunenwald,  MD,  and Muriel Fartoukh,  MD Respiratory and Critical Care, Thoracic Surgery, and Radiology Department, Tenon Hospital, Assistance Publique - Hôpitaux deParis and Pierre et Marie Curie University and INSERM U 149, Paris, France Background.  The role of surgical lung resection in themanagement of severe hemoptysis has evolved afteradvances in interventional radiology. We sought to de-scribe the indications for surgical lung resection in suchpatients and to identify predictive factors of postopera-tive complications.  Methods.  This study is a retrospective analysis (May1995 to July 2006) of consecutive patients referred to theintensive care unit of a tertiary hospital for severe he-moptysis who underwent surgical lung resection. Results.  Among 813 patients referred for severe he-moptysis, 111 underwent surgical lung resection. Inter-ventional radiology had been first attempted in 87 pa-tients (78%); 68 underwent surgery because of a failedprocedure (n  28) or bleeding persistence or recurrencewithin 72 hours despite a completed procedure (n  40);19 patients underwent surgery after bleeding control.The remaining 24 patients (22%) were directly referred tothe surgeon (5 for emergency surgery). Overall, surgerywas performed in emergency (n    48), scheduled afterbleeding control (n  48), or planned after discharge (n  15). The main indications for surgery were mycetoma,cancer, bronchiectasis, and active tuberculosis. Surgeryformycetoma(oddsratio,9.4;95%confidenceinterval,2.8to 32), emergency surgery (odds ratio, 5.3; 95% confidenceinterval, 1.8 to 16), and pneumonectomy (odds ratio, 4.7;95% confidence interval, 1.2 to 18) independently pre-dicted complications. Fifteen patients died in the inten-sive care unit, of whom 14 underwent emergency sur-gery. Chronic alcoholism (odds ratio, 4.6; 95% confidenceinterval, 1.1 to 19), the need for mechanical ventilation orvasoactive drugs on admission (odds ratio, 8.2; 95%confidence interval, 1.9 to 35), and blood transfusionbefore surgery (odds ratio, 8; 95% confidence interval, 1.5to 42) predicted mortality. Conclusions.  Attempting at controlling bleeding withfirst-line nonsurgical approaches appears necessary tooptimize the operative conditions and improve outcomeof patients with severe hemoptysis.(Ann Thorac Surg 2009;88:1556–65)© 2009 by The Society of Thoracic Surgeons S evere hemoptysis (SH) is a life-threatening conditionassociated with a high mortality rate in the absenceof adequate treatment. Until two decades ago, surgicallung resection was considered as the treatment of choicefor SH in patients with adequate pulmonary function,with the only alternative being general supportive care[1, 2]. Surgical lung resection for SH was also reputed tobe associated with high morbidity and mortality, espe-cially when performed during active bleeding [3–6]. The management of SH has largely evolved since the firstreports of successful bronchial artery embolization (BAE)[7] , now considered as first-line therapy [8, 9]. Further- more, more advanced and promising interventional radiol-ogy procedures have been recently developed such aspulmonary angiography with vasoocclusion [10]. However, no controlled study has compared the medical (includinginterventional radiology) versus the surgical approach inthe initial management of SH. Even if BAE, and moregenerally interventional radiology, has become widelyused, a subset of patients with SH still requires surgicallung resection at the early stage of their management.In this study, we analyze a series of consecutive patientswith SH referred to a single intensive care unit (ICU) withan affiliated step-down unit (respectively, 10 and 6 beds)who underwent surgical lung resection. These patientswere selected from a large cohort of 813 patients referred toour unit for SH. Our aims were (1) to identify factorsassociated with surgery among patients with SH; (2) todescribe the clinical epidemiology of patients referred tosurgery and the conditions in which surgery was per-formed; and (3) to identify factors associated with postop-erative complications and ICU mortality, accounting forwhether bleeding was controlled or not at time of surgery.The study was conducted in accordance with French law,which does not require approval of an institutional review Accepted for publication June 4, 2009.Address correspondence to Dr Fartoukh, Hôpital Tenon, AssistancePublique, Hôpitaux de Paris, 04 rue de la Chine, Paris, 75020, France;e-mail: muriel.fartoukh@tnn.aphp.fr. © 2009 by The Society of Thoracic Surgeons 0003-4975/09/$36.00Published by Elsevier Inc doi:10.1016/j.athoracsur.2009.06.011  G E  N E  R A L  T  H O R A C I   C  board or individual patients’ consent for such retrospectiveanalysis of medical records. Patients and Methods  Patients The study was conducted during an 11-year period (May1995 to July 2006) at Tenon Hospital, a tertiary universityhospital and referral center for hemoptysis in Paris,France. All consecutive patients admitted to our institu-tion with SH who underwent surgical lung resectionwere eligible. Patients with hemoptysis who underwent adiagnostic open lung biopsy were excluded. For eachpatient, baseline demographics, comorbid conditions,initial clinical presentation and vital signs, cause of hemoptysis, management, indications and conditions of surgery, ICU and hospital lengths of stay, and vital statusat discharge were recorded. Definitions SEVERITY OF HEMOPTYSIS.  One of the following conditionsdefined SH: (1) a cumulative amount of bleeding on admis-sionofatleast200mLwithinthepreceding72hoursincaseof normal or nearly normal lung function, (2) acute respi-ratory failure, (3) use of intravenous vasoconstrictive agent(terlipressin), and (4) need for blood transfusions. Thecumulative amount of bleeding was assessed from theonset of bleeding until the first hours of admission to ourunit using the following scale: a spoonful (5 mL), a smallfilled glass (100 mL), and a large filled glass (200 mL). Acuterespiratory failure was defined as room air oxygen satura-tion less than 90% or a respiratory rate greater than 25breaths/min, and the requirement for high-flow oxygen ormechanical ventilation. CAUSE OF HEMOPTYSIS.  The cause of hemoptysis was diag-nosed on the combination of detailed medical history,physical examination, chest roentgenograph, fiberopticbronchoscopy, computed tomography scan, microbio-logic results if performed, and pathologic examination of lung tissue resected.  Management  The main objective of the initial management was tocontrol the bleeding to prevent asphyxiation. The pa-tients received general supportive care in a uniform way,as described elsewhere [11]. The severity of bleeding was assessed on the need for administration of local orsystemic vasopressors (terlipressin), mechanical ventila-tion, or blood transfusions or vasoactive drugs (epineph-rine, norepinephrine) to control hypotension, before re-ferral or within the first 24 hours of ICU admission.Intrabronchial procedures used in an attempt to controlthe bleeding combined the bronchoscopic topical tech-niques (blood aspiration, local instillation of cold salinelavage and of vasopressors if needed), the bronchoscopicplacement of a balloon, the administration of intravenousterlipressin, and the use of a double-lumen endotrachealtube or a selective intubation.A bronchial arteriography was first attempted for BAEin combination with general supportive care. In case of an unstable catheter or dangerous collaterals, a superse-lective catheterization was performed using a microcath-eter, as recommended [12]. A pulmonary angiography, if  possible associated with pulmonary vasoocclusion, wasperformed alone in patients with hemoptysis suspectedof a pulmonary arterial mechanism, or in association withBAE in patients with persistent or recurrent bleeding.Failure of attempted bronchial arteriography refers to thepersistence of bleeding in the following situations: nosystemic hypervascularization, inability to cannulate thevessel, or instability of the catheter tip despite superse-lective catheterization. Failure of the attempted pulmo-nary angiography refers to the lack of demonstration of apulmonary arterial source for hemoptysis. Failure of completed bronchial arteriography (or pulmonary an-giography) refers to the persistence or the recurrence of bleeding despite embolization (or vasoocclusion).The timing of surgery was defined as emergency sur-gery, scheduled, or planned. Emergency surgical resec-tion was performed during active and uncontrolledbleeding. When performed after control of bleeding,surgery was either scheduled during the same hospital-ization, or secondarily planned (within 6 months of discharge), because of an expected high risk of recur-rence of bleeding. Outcome Intraoperative complications included the followingevents occurring during surgery: acute hemorrhagic ane-mia (defined as the need for blood products transfusion),shock (defined as the need for intravenous infusion of vasopressors such as epinephrine or norepinephrine),and death.Postoperative complications were defined as complica-tions occurring 24 hours or more after surgery andincluded (1) the need for mechanical ventilation; (2) theneed for blood products transfusion; (3) the need forinfusion of vasopressor drugs (epinephrine, norepineph-rine) to maintain hemodynamic stability; (4) the occur-rence of a bronchopleural fistula, a pneumonia or anyextrapulmonary hospital-acquired infection; and (5)death.Intensive care unit and hospital lengths of stay andvital status at discharge were recorded. The follow-upwas conducted during a visit in our institution or by atelephone interview after discharge. Statistical Analysis Our study first aimed at analyzing the factors associatedwith performing surgical lung resection and describing theclinical epidemiology of patients with SH who underwentsurgery. The patients’ demographics and clinical variableswereanalyzed,usingusualdescriptivestatistics.Resultsareexpressed as mean    standard deviation (range), unlessotherwise stated. The Mann-Whitney  U   test was used forquantitative variables and the  2 test or Fisher’s exact testfornominalvariables.Aprobabilityvaluelessthan0.05wasconsidered statistically significant.Second, we identified predictive factors of postopera-tive complications and of ICU mortality by performing1557 Ann Thorac Surg ANDRÉJAK ET AL2009;88:1556–65 SURGICAL LUNG RESECTION FOR HEMOPTYSIS      G     E     N     E     R     A     L     T     H     O     R     A     C     I     C  multivariate analyses using a stepwise backward logisticregression. The number of events per variable entered inthe final multivariate model averaged a ratio of 10 toavoid overfitting, as recommended [13]. The simplified acute physiology score (SAPS II) was not entered in bothmultivariate analyses, because it encompassed numerousvariables used to define organ failures that overlappedwith other variables considered in the models. Table 1. Characteristics of 779 Patients With Hemoptysis, Contrasting Medical and Surgical Patients CharacteristicSurgical Population(n  111)Medical Population(n  668)  p  ValueAge (y), mean  SD (range) 50  14 (16–79) 55  18 (16–92) 0.003Males, n (%) 82 (74) 488 (73) NSPast history of hemoptysis, n (%) 48 (43) 201 (30) 0.01Cumulated volume of blood on admission (mL), mean  SD (range) 280  220 (10–1000) 210  190 (10–1200)   0.001Cumulated volume of blood  200 mL on admission, n (%) 66 (62) 284 (44)   0.001Mechanical ventilation on admission, n (%) 17 (15) 40 (6) 0.001 Vasoactive drugs on admission, n (%) 14 (13) 13 (2)   0.001Failed first-line interventional radiology, n (%) 68/87 (78) 119/452 (26)   0.001Cause of hemoptysis, n (%) a Mycetoma 28 (25) 17 (3)   0.001Cancer 24 (22) 80 (13) 0.02Bronchiectasis 21 (19) 242 (38)   0.001Cryptogenic 12 (11) 105 (17) 0.2Active tuberculosis 8 (7) 94 (15) 0.04Pneumonia 6 (5) 19 (3) 0.3Pulmonary arterial aneurysm 4 (4) 3 (1) 0.035Others 10 (9) 82 (13) 0.3 a Mycetoma and active tuberculosis were both evidenced on pathologic examination of the lung resection in 2 surgical patients. Two causes wereidentified in 12 medical patients.SD  standard deviation. Table 2. Clinical Characteristics of the Surgical Population a CharacteristicSurgical Population(n  111)Emergency SurgicalResection (n  48)Scheduled SurgicalResection (n  48)Planned SurgicalResection (n  15)Age (y), mean  SD (range) 50  14 (16–79) 49  14 (19–76) 50  12 (21–70) 53  17 (16–79)Males, n (%) 82 (74) 39 (81) 32 (67) 11 (73)Cumulated volume of blood on admission(mL), mean  SD (range)275  223 (10–1000) 310  240 (40–1000) 260  190 (30–800) 230  290 (10–1000)Tobacco use, n (%) 67 (60) 26 (54) 30 (63) 11 (73)Chronic alcoholism, n (%) 37 (33) 19 (40) 16 (33) 2 (13)Comorbid conditions, n (%)Extrapulmonary, n (%) 56 (50) 31 (65) 16 (33) 9 (60)Cardiovascular 20 (18) 9 (19) 5 (10) 6 (40)Cancer (except lung) 16 (14) 7 (15) 7 (15) 2 (13)Hepatic disease 7 (6) 6 (13) 0 1 (7)HIV 4 (4) 3 (6) 1 (2) 0Others 9 (8) 6 (13) 3 (6) 0Pulmonary, n (%)Tuberculosis 65 (59) 24 (50) 32 (67) 9 (60)Chronic 33 (30) 14 (29) 16 (33) 3 (20)bronchitis/COPD 15 (14) 4 (8) 8 (17) 3 (20)Bronchiectasis 6 (5) 1 (2) 3 (6) 2 (13)Cancer 5 (5) 2 (4) 2 (4) 1 (7)Others 6 (5) 3 (6) 3 (6) 0 a Results are expressed as mean  SD (range), unless otherwise stated.COPD  chronic obstructive pulmonary disease; HIV  human immunodeficiency virus; SD  standard deviation. 1558  ANDRÉJAK ET AL Ann Thorac SurgSURGICAL LUNG RESECTION FOR HEMOPTYSIS 2009;88:1556–65  G E  N E  R A L  T  H O R A C I   C  Univariate analysis first assessed the association be-tween each variable and postoperative complications. Variables selected by univariate analysis (  p    0.2) andthose considered clinically relevant (tobacco exposure,mechanical ventilation or vasoactive drugs on admission,blood transfusion, cause of hemoptysis, conditions of surgery, and surgical procedure) were entered in a logis-tic regression model to identify the predictors of postop-erative complications. For the analysis of ICU mortality,clinically relevant variables (chronic alcoholism, mechan-ical ventilation or vasoactive drugs on admission, andblood transfusion) selected by univariate analysis wereentered in a logistic regression model to identify predic-tors of ICU mortality. Emergency surgery was not en-tered in this model, because most patients (n  14 of 15;93%) who died in the ICU had undergone emergencysurgical resection. The results are reported using oddsratio (OR) and corresponding 95% confidence intervals(CI). All statistical analyses were performed on a per-sonal computer using the Statview software (SAS Insti-tute, Cary, NC). Results During the 11-year study period, 813 consecutive patientswere referred to our unit for SH, of whom 116 (14%)underwent a surgical lung resection. Five patients wereexcluded from the analysis (diagnostic surgical lungbiopsy, n  2; surgery planned 6 months after the initialepisode of hemoptysis, n  2; and surgery performed atanother center, n    1). Overall, 111 patients undergoingsurgical resection for SH in our center were analyzed.Some of these patients have been described elsewhere[11, 14, 15]. Twenty-nine medical patients had missingdata and were excluded. The characteristics of patientsmanaged with general supportive care and interven-tional radiology only (n    668) or undergoing surgery(n  111) are shown in Table 1. Variables associated with performing surgery were related to the magnitude andconsequences of bleeding (volume of bleeding, mechan-ical ventilation, or shock), the noncontrol of bleedingusing general supportive care and interventional radiol-ogy, or an expected high risk of bleeding recurrencerelated to the cause or the mechanism of hemoptysis.Patients undergoing surgery (82 males) were aged 50  14 years. Most of them were referred from anotherhospital (n  74; 67%). The mean time between bleedingonset and referral to our unit was 3    5 days (mediantime, 1 day; Table 2). Forty-nine patients (44%) received intravenous terlipressin before referral (n    36), or dur-ing the first 24 hours of admission (n    13). Sixty-onepatients (55%) presented with acute respiratory failure,and 15 required mechanical ventilation within the first 24hours. Blood transfusions were administered to 21 pa-tients (19%) during the first 24 hours and overall to 40patients (36%) before surgery. Vasoactive drugs (epi-nephrine, norepinephrine) were administered to 14 pa-tients (13%) within the first 24 hours of ICU admission.Bronchoscopic techniques, combining blood aspirationand local instillation of cold saline solution, were per-formed in 57 patients (51%). Terlipressin or adrenalinewas bronchoscopically delivered in 47 patients (42%), anda balloon was placed in 6 (12%). The main indications forsurgery were mycetoma (25%), cancer (22%), bronchiec-tasis (19%), and tuberculosis (7%) (Table 1). Twenty-four patients (22%) were referred directly tothe surgeon for lung resection, ie, without first-lineattempt to undergo interventional radiology. The indica-tions for direct surgical referral included an expectedtechnically hazardous procedure of BAE (n  2), a recenthistory of massive bleeding treated with a completedprocedure of interventional radiology (n    5), involve-ment of the proximal region of a pulmonary artery (n  2), the associated need for a pathologic examination for Table 3. Details of Circumstances Leading to Surgical Lung Resection in 111 Patients  VariableEmergency SurgicalLung Resection(n  48)Scheduled SurgicalLung Resection(n  48)Planned SurgicalLung Resection(n  15)General supportive care alone, n (%)Failure 5 (10) 0 0Success 0 11 (23) 8 (53)First-line interventional radiology, n (%)Failure 16 (33) 6 (12.5) 1 (7)Success 0 4 (8) 1 (7)First-line completed interventional radiology, n (%)Failure 27 (56) 12 (25) 1 (7)Success 0 15 (31) 4 (27)Cumulative volume at surgical lung resection (mL) 700  440 (150–1800) 340  280 (30–1200) 290  340 (10–1100)Time of surgical lung resection after admission (days) 5  6 (0–22) 8.5  7 (1–27) 61  49 (8–178)Type of surgical lung resection, n (%)Pneumonectomy 11 (23) 8 (17) 2 (13)Lobectomy 30 (63) 36 (75) 12 (80)Segmentectomy 4 (8) 4 (8) 1 (7) 1559 Ann Thorac Surg ANDRÉJAK ET AL2009;88:1556–65 SURGICAL LUNG RESECTION FOR HEMOPTYSIS      G     E     N     E     R     A     L     T     H     O     R     A     C     I     C
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks