Thoracic aortic endovascular repair for mycotic aneurysms and fistulas
2010; Elsevier BV; Volume: 52; Issue: 4 Linguagem: Inglês
10.1016/j.jvs.2010.06.139
ISSN1097-6809
AutoresHimanshu J. Patel, David M. Williams, Gilbert R. Upchurch, Narasimham L. Dasika, Jonathan L. Eliason, G. Michael Deeb,
Tópico(s)Aortic Disease and Treatment Approaches
ResumoInfected thoracic aortic aneurysms are a rare pathologic entity.1Hsu R.B. Lin F.Y. Infected aneurysm of the thoracic aorta.J Vasc Surg. 2008; 47: 270-276Abstract Full Text Full Text PDF PubMed Scopus (83) Google Scholar, 2Muller B.T. Wegener O.R. Grabitz K. Pillny M. Thomas L. Sandmann W. Mycotic aneurysms of the thoracic and abdominal aorta and iliac arteries: experience with anatomic and extra-anatomic repair in 33 cases.J Vasc Surg. 2001; 33: 106-113Abstract Full Text Full Text PDF PubMed Scopus (481) Google Scholar, 3Hsu R.B. Lin F.Y. Surgery for infected aneurysm of the aortic arch.J Thorac Cardiovasc Surg. 2007; 134: 1157-1162Abstract Full Text Full Text PDF PubMed Scopus (31) Google Scholar, 4von Segesser L.K. Tkebuchava T. Niederhauser U. Kunzli A. Lachat M. Genoni M. et al.Aortobronchial and aortoesophageal fistulae as risk factors in surgery of descending thoracic aortic aneurysms.Eur J Cardiothorac Surg. 1997; 12: 195-201Crossref PubMed Scopus (50) Google Scholar, 5Patel H.J. Williams D.W. Upchurch G.R. Dasika N.L. Eliason J.L. Deeb G.M. et al.Late outcomes of endovascular repair for the infected thoracic aorta.Ann Thorac Surg. 2009; 87: 1366-1371Abstract Full Text Full Text PDF PubMed Scopus (36) Google Scholar, 6Ting A.C. Cheng S.W. Ho P. Chan Y.C. Poon J.T. Cheung G.C. Endovascular stent graft repair for infected thoracic aortic pseudoaneurysms—a durable option?.J Vasc Surg. 2006; 44: 701-705Abstract Full Text Full Text PDF PubMed Scopus (73) Google Scholar, 7Gonzalez-Fajardo J.A. Gutierrez V. Martin-Pedrosa M. Del Rio L. Carrera S. Vaquero C. Endovascular repair in the presence of aortic infection.Ann Vasc Surg. 2005; 19: 94-98Abstract Full Text Full Text PDF PubMed Scopus (77) Google Scholar, 8Wheatley G.H. Nunez A. Preventza O. Ramaiah V.G. Rodriguez-Lopez J.A. Williams J. et al.Have we gone too far? Endovascular stent-graft repair of aortobronchial fistulas.J Thorac Cardiovasc Surg. 2007; 133: 1277-1285Abstract Full Text Full Text PDF PubMed Scopus (46) Google Scholar, 9Malouf J.F. Chandrasekaran K. Orszulak T.A. Mycotic aneurysms of the thoracic aorta: a diagnostic challenge.Am J Med. 2003; 115: 489-496Abstract Full Text Full Text PDF PubMed Scopus (76) Google Scholar, 10Oderich G.S. Panneton J.M. Bower T.C. Cherry Jr, K.J. Rowland C.M. Noel A.A. et al.Infected aortic aneurysms: aggressive presentation, complicated early outcome, but durable results.J Vasc Surg. 2001; 34: 900-908Abstract Full Text Full Text PDF PubMed Scopus (314) Google Scholar, 11Reisenmann P.J. Books J.D. Farber M.A. Thoracic endovascular repair of aortobronchial fistulas.J Vasc Surg. 2009; 50: 992-998Abstract Full Text Full Text PDF PubMed Scopus (38) Google Scholar Presentation can occur with constitutional symptoms (fevers, failure to thrive, weight loss), chest pain, or with symptoms attributable to adjacent organ development. The latter group often includes back pain from erosion into vertebrae, hemoptysis from aortobronchial fistulae, and hematemesis or odynophagia from aortoesophageal fistulae. The diagnosis is made based on clinical presentation, characteristic radiographic findings, and microbiologic data. The underlying aortic pathology in mycotic aneurysm is typically a saccular configuration. In contrast, fistulae between the thoracic aorta and adjacent organ systems can present with underlying fusiform, dissecting, or saccular aneurysms. The mortality of medically treated infected thoracic aortic pathology (ITAP) exceeds 60%.1Hsu R.B. Lin F.Y. Infected aneurysm of the thoracic aorta.J Vasc Surg. 2008; 47: 270-276Abstract Full Text Full Text PDF PubMed Scopus (83) Google Scholar While open surgical repair has been considered the gold standard, reported mortality from this approach can exceed 12% in the largest series to date.1Hsu R.B. Lin F.Y. Infected aneurysm of the thoracic aorta.J Vasc Surg. 2008; 47: 270-276Abstract Full Text Full Text PDF PubMed Scopus (83) Google Scholar, 2Muller B.T. Wegener O.R. Grabitz K. Pillny M. Thomas L. Sandmann W. Mycotic aneurysms of the thoracic and abdominal aorta and iliac arteries: experience with anatomic and extra-anatomic repair in 33 cases.J Vasc Surg. 2001; 33: 106-113Abstract Full Text Full Text PDF PubMed Scopus (481) Google Scholar, 3Hsu R.B. Lin F.Y. Surgery for infected aneurysm of the aortic arch.J Thorac Cardiovasc Surg. 2007; 134: 1157-1162Abstract Full Text Full Text PDF PubMed Scopus (31) Google Scholar, 4von Segesser L.K. Tkebuchava T. Niederhauser U. Kunzli A. Lachat M. Genoni M. et al.Aortobronchial and aortoesophageal fistulae as risk factors in surgery of descending thoracic aortic aneurysms.Eur J Cardiothorac Surg. 1997; 12: 195-201Crossref PubMed Scopus (50) Google Scholar Classic surgical principles in the treatment of open repair include wide debridement of infected tissue and extra-anatomic bypass. Recent work has suggested that in situ grafting may be appropriate, and some authors have recognized the potential for use of rifampin-soaked grafts as a means for reducing risk for reinfection.1Hsu R.B. Lin F.Y. Infected aneurysm of the thoracic aorta.J Vasc Surg. 2008; 47: 270-276Abstract Full Text Full Text PDF PubMed Scopus (83) Google Scholar Recognizing the potential role of endovascular therapy (TEVAR), we and others have suggested that TEVAR represents a minimally invasive option in treating often moribund patients and have reported early mortality rates as low as 0%.5Patel H.J. Williams D.W. Upchurch G.R. Dasika N.L. Eliason J.L. Deeb G.M. et al.Late outcomes of endovascular repair for the infected thoracic aorta.Ann Thorac Surg. 2009; 87: 1366-1371Abstract Full Text Full Text PDF PubMed Scopus (36) Google Scholar, 6Ting A.C. Cheng S.W. Ho P. Chan Y.C. Poon J.T. Cheung G.C. Endovascular stent graft repair for infected thoracic aortic pseudoaneurysms—a durable option?.J Vasc Surg. 2006; 44: 701-705Abstract Full Text Full Text PDF PubMed Scopus (73) Google Scholar, 7Gonzalez-Fajardo J.A. Gutierrez V. Martin-Pedrosa M. Del Rio L. Carrera S. Vaquero C. Endovascular repair in the presence of aortic infection.Ann Vasc Surg. 2005; 19: 94-98Abstract Full Text Full Text PDF PubMed Scopus (77) Google Scholar, 8Wheatley G.H. Nunez A. Preventza O. Ramaiah V.G. Rodriguez-Lopez J.A. Williams J. et al.Have we gone too far? Endovascular stent-graft repair of aortobronchial fistulas.J Thorac Cardiovasc Surg. 2007; 133: 1277-1285Abstract Full Text Full Text PDF PubMed Scopus (46) Google Scholar, 11Reisenmann P.J. Books J.D. Farber M.A. Thoracic endovascular repair of aortobronchial fistulas.J Vasc Surg. 2009; 50: 992-998Abstract Full Text Full Text PDF PubMed Scopus (38) Google Scholar This review updates our previous work and summarizes the current efforts reported in the literature. This retrospective study was approved by the Institutional Review Board (IRB) of the University of Michigan Hospitals. Informed consent requirements were waived for this study. From 1993 to 2010, 27 patients (median age 72 years, 66.7% male) have undergone TEVAR for infected thoracic aortic pathology at the University of Michigan Hospitals. Of this group, 26 were considered high risk for open surgery, while the remaining patient refused open surgery and was offered TEVAR as an alternative. In six patients (22.2%), the procedure was performed in the setting of an infected Dacron graft. The underlying pathology contributing to the presentation included saccular aneurysm (23), aortic dissection (2), and fusiform aneurysm (1), while the remaining patient developed a fistula between an otherwise normal aortic contour and a gastric conduit used after transhiatal esophagectomy. Of the entire study cohort, 14 patients presented with a mycotic aneurysm. Sixteen patients from the entire cohort had associated fistulous connections, including aortobronchial (12), aortoesophageal (3), and aortocutaneous (1) fistulae. Statistical methods to analyze the data included univariate (χ2 or independent sample t test) and multivariate (Cox proportional hazards analysis with forward selection process). Survival was analyzed by Kaplan-Meier analysis and vital status verified using the Social Security Death Index. Follow-up for the primary outcome of all cause mortality was 100% complete (mean 26.4 ± 33.6 months). Comorbidities in this group were prevalent and included hypertension (18), chronic obstructive pulmonary disease (7), coronary artery disease (8), peripheral vascular disease (6), and a history of tobacco use (19). One presented with end-stage renal failure needing dialysis, and the mean creatinine for the cohort was 1.4 ± 1.5 mg/dL. Importantly, a history of immunosuppression was identified in four and malignancy in seven. Prior aortic repair in this cohort consisted of abdominal aortic aneurysmectomy (4), prior descending aortic repair (3), and aortic valve replacement with ascending aneurysmectomy (1). Technical success was achieved in all but one patient. That patient presented with a graft infection following two prior thoracotomies and for coarctation repair and then repair of a large postcoarctation pseudoaneurysm. He had an inadequate proximal landing zone to extend therapy to exclude the sac and underwent uneventful graft explantation and extra-anatomic ascending to supraceliac aortic bypass with left carotid artery debranching the next day. Concomitant procedures included hybrid arch debranching with (1) and without coronary artery bypass grafting (1), and visceral/renal debranching (2). In-hospital mortality was seen in three patients (11.5%). The causes of mortality included refractory hypoxemia following massive hemoptysis prior to TEVAR in one, multidrug resistant pneumonia in another, and finally tracheoesophageal fistula in one who presented with an aortoesophageal fistula. Major morbidity included the need for hemodialysis in two patients (7.4%), both of whom had impaired renal function at baseline (creatinine ≥2.0 mg/dL). The incidence of postoperative pneumonia was five (18.5%). Neither stroke nor spinal cord ischemia was identified following TEVAR. The antibiotic regimen in this group varied but included the institution of antibiotics for at least 48 hours prior to surgery in 18 (66.7%). Postoperative antibiotic regimens included prolonged courses (greater than 3 weeks) in all but one patient. This patient presented with hemoptysis; the aortobronchial fistula was initially suspected secondary to pulmonary parenchymal erosion rather than infection. She died 4 months later, following identification of an endoleak and gas bubbles in the aneurysm sac consistent with infection. At the age of 95 years, she was not considered a suitable candidate for open graft explantation and extra-anatomic bypass. In this debilitated population, the overall crude mortality rate was 65% (n = 14), with four deaths attributable to aortic reinfection. The Kaplan-Meier survival curve is shown in Fig 1 and demonstrates a 3-year survival of 57.8%. There were no univariate correlates with late mortality. Endoleak was seen in five patients (18.5%). Proximal landing zone endoleaks were identified in one patient occurring intraoperatively from inadequate proximal landing zone length and described above. The remaining patients had indeterminate endoleaks. One underwent TEVAR for an aortobronchial fistula in the setting of previous chronic dissection. He continues to be followed, but has a stable indeterminate endoleak (type 2 or 3) with a decreasing sac size 2 years following TEVAR. An indeterminate type was identified in another who eventually expired 4 months after TEVAR of unknown cause. The third patient (95-year-old) developed an indeterminate endoleak (type 1 or 2) with gas bubbles in the aneurysm sac 4 months after TEVAR as described above. The last patient is currently being followed and has a stable sac size. At 2 years, Kaplan-Meier freedom from reinfection was 81.2% (Fig 2,A). Univariate variables correlating with reinfection of the endograft included initial diagnosis of fistulous connection (vs mycotic aneurysm, P = .06), TEVAR extending into the arch (P = .047) or for prosthetic graft infection (vs native graft, P = .004). When these variables were entered into Cox multivariate analysis, only TEVAR for a prosthetic graft infection was independently associated with an increased risk for reinfection (HR = 8.05, P = .063). This was confirmed following Kaplan-Meier analysis, where TEVAR into the native aorta portended a decreased risk for reinfection (Fig 2, B, log rank P = .028). The infected thoracic aorta remains among the most formidable of clinical challenges. While optimal medical therapy is associated with a dismal prognosis, any operative intervention is associated with high rates of morbidity and mortality.1Hsu R.B. Lin F.Y. Infected aneurysm of the thoracic aorta.J Vasc Surg. 2008; 47: 270-276Abstract Full Text Full Text PDF PubMed Scopus (83) Google Scholar, 2Muller B.T. Wegener O.R. Grabitz K. Pillny M. Thomas L. Sandmann W. Mycotic aneurysms of the thoracic and abdominal aorta and iliac arteries: experience with anatomic and extra-anatomic repair in 33 cases.J Vasc Surg. 2001; 33: 106-113Abstract Full Text Full Text PDF PubMed Scopus (481) Google Scholar, 3Hsu R.B. Lin F.Y. Surgery for infected aneurysm of the aortic arch.J Thorac Cardiovasc Surg. 2007; 134: 1157-1162Abstract Full Text Full Text PDF PubMed Scopus (31) Google Scholar, 4von Segesser L.K. Tkebuchava T. Niederhauser U. Kunzli A. Lachat M. Genoni M. et al.Aortobronchial and aortoesophageal fistulae as risk factors in surgery of descending thoracic aortic aneurysms.Eur J Cardiothorac Surg. 1997; 12: 195-201Crossref PubMed Scopus (50) Google Scholar, 5Patel H.J. Williams D.W. Upchurch G.R. Dasika N.L. Eliason J.L. Deeb G.M. et al.Late outcomes of endovascular repair for the infected thoracic aorta.Ann Thorac Surg. 2009; 87: 1366-1371Abstract Full Text Full Text PDF PubMed Scopus (36) Google Scholar, 6Ting A.C. Cheng S.W. Ho P. Chan Y.C. Poon J.T. Cheung G.C. Endovascular stent graft repair for infected thoracic aortic pseudoaneurysms—a durable option?.J Vasc Surg. 2006; 44: 701-705Abstract Full Text Full Text PDF PubMed Scopus (73) Google Scholar, 7Gonzalez-Fajardo J.A. Gutierrez V. Martin-Pedrosa M. Del Rio L. Carrera S. Vaquero C. Endovascular repair in the presence of aortic infection.Ann Vasc Surg. 2005; 19: 94-98Abstract Full Text Full Text PDF PubMed Scopus (77) Google Scholar, 8Wheatley G.H. Nunez A. Preventza O. Ramaiah V.G. Rodriguez-Lopez J.A. Williams J. et al.Have we gone too far? Endovascular stent-graft repair of aortobronchial fistulas.J Thorac Cardiovasc Surg. 2007; 133: 1277-1285Abstract Full Text Full Text PDF PubMed Scopus (46) Google Scholar, 9Malouf J.F. Chandrasekaran K. Orszulak T.A. Mycotic aneurysms of the thoracic aorta: a diagnostic challenge.Am J Med. 2003; 115: 489-496Abstract Full Text Full Text PDF PubMed Scopus (76) Google Scholar, 10Oderich G.S. Panneton J.M. Bower T.C. Cherry Jr, K.J. Rowland C.M. Noel A.A. et al.Infected aortic aneurysms: aggressive presentation, complicated early outcome, but durable results.J Vasc Surg. 2001; 34: 900-908Abstract Full Text Full Text PDF PubMed Scopus (314) Google Scholar, 11Reisenmann P.J. Books J.D. Farber M.A. Thoracic endovascular repair of aortobronchial fistulas.J Vasc Surg. 2009; 50: 992-998Abstract Full Text Full Text PDF PubMed Scopus (38) Google Scholar The advent of thoracic endovascular repair as a less invasive therapy has the potential to resolve this judgmental paradox. Previous authors have reported their experiences with early and late results of TEVAR.5Patel H.J. Williams D.W. Upchurch G.R. Dasika N.L. Eliason J.L. Deeb G.M. et al.Late outcomes of endovascular repair for the infected thoracic aorta.Ann Thorac Surg. 2009; 87: 1366-1371Abstract Full Text Full Text PDF PubMed Scopus (36) Google Scholar, 6Ting A.C. Cheng S.W. Ho P. Chan Y.C. Poon J.T. Cheung G.C. Endovascular stent graft repair for infected thoracic aortic pseudoaneurysms—a durable option?.J Vasc Surg. 2006; 44: 701-705Abstract Full Text Full Text PDF PubMed Scopus (73) Google Scholar, 7Gonzalez-Fajardo J.A. Gutierrez V. Martin-Pedrosa M. Del Rio L. Carrera S. Vaquero C. Endovascular repair in the presence of aortic infection.Ann Vasc Surg. 2005; 19: 94-98Abstract Full Text Full Text PDF PubMed Scopus (77) Google Scholar, 8Wheatley G.H. Nunez A. Preventza O. Ramaiah V.G. Rodriguez-Lopez J.A. Williams J. et al.Have we gone too far? Endovascular stent-graft repair of aortobronchial fistulas.J Thorac Cardiovasc Surg. 2007; 133: 1277-1285Abstract Full Text Full Text PDF PubMed Scopus (46) Google Scholar, 11Reisenmann P.J. Books J.D. Farber M.A. Thoracic endovascular repair of aortobronchial fistulas.J Vasc Surg. 2009; 50: 992-998Abstract Full Text Full Text PDF PubMed Scopus (38) Google Scholar, 12Bockler D. Schumacher H. Schwarzbach M. Ockert S. Rotert H. Allenberg J.R. Endoluminal stent-graft repair of aortobronchial fistulas: bridging or definitive long-term solution.J Endovasc Ther. 2004; 11: 41-48Crossref PubMed Scopus (39) Google Scholar, 13Perrelli S. Bozzani A. Arici V. Odero A. Endovascular treatment of acute hemoptisis secondary to aortobronchial fistula.Eur J Vasc Endovasc Surg. 2006; 32: 366-368Abstract Full Text Full Text PDF PubMed Scopus (35) Google Scholar However, these outcomes have been mixed, with some authors suggesting a low rate of reinfection. A summary of recent work is listed in the Table. Heterogeneous study groups described in these analyses may explain the discrepant results listed. For example, aortobronchial fistulae can occur secondary to erosion of the aneurysm into pulmonary parenchymal bronchioles rather than to a pneumonia infecting an aneurysm or prosthetic graft. The results of TEVAR in this particular example are likely to be different. Different duration of antibiotic therapy may also explain these differences in outcomes.5Patel H.J. Williams D.W. Upchurch G.R. Dasika N.L. Eliason J.L. Deeb G.M. et al.Late outcomes of endovascular repair for the infected thoracic aorta.Ann Thorac Surg. 2009; 87: 1366-1371Abstract Full Text Full Text PDF PubMed Scopus (36) Google Scholar, 7Gonzalez-Fajardo J.A. Gutierrez V. Martin-Pedrosa M. Del Rio L. Carrera S. Vaquero C. Endovascular repair in the presence of aortic infection.Ann Vasc Surg. 2005; 19: 94-98Abstract Full Text Full Text PDF PubMed Scopus (77) Google Scholar, 8Wheatley G.H. Nunez A. Preventza O. Ramaiah V.G. Rodriguez-Lopez J.A. Williams J. et al.Have we gone too far? Endovascular stent-graft repair of aortobronchial fistulas.J Thorac Cardiovasc Surg. 2007; 133: 1277-1285Abstract Full Text Full Text PDF PubMed Scopus (46) Google Scholar Our preference is to recommend antibiotics for at least 6 weeks, and preferentially for life in this setting where wide debridement of infected tissues is not performed and endografts are maintained in an infected field. An analogy is drawn to the treatment of endocarditis where extensive infection is treated with prolonged courses of antibiotics and the source of infection is removed.TableSummary of series treating infectious thoracic aortic pathology with endovascular repairFirst authorYear of studyPresentationEarly mortalityLate reinfectionBockler12Bockler D. Schumacher H. Schwarzbach M. Ockert S. Rotert H. Allenberg J.R. Endoluminal stent-graft repair of aortobronchial fistulas: bridging or definitive long-term solution.J Endovasc Ther. 2004; 11: 41-48Crossref PubMed Scopus (39) Google Scholar2004Aortobronchial fistula01 (13%)Gonzalez-Fajardo7Gonzalez-Fajardo J.A. Gutierrez V. Martin-Pedrosa M. Del Rio L. Carrera S. Vaquero C. Endovascular repair in the presence of aortic infection.Ann Vasc Surg. 2005; 19: 94-98Abstract Full Text Full Text PDF PubMed Scopus (77) Google Scholar2005Aortoesophageal fistula1 (50%)2 (100%)Pirrelli13Perrelli S. Bozzani A. Arici V. Odero A. Endovascular treatment of acute hemoptisis secondary to aortobronchial fistula.Eur J Vasc Endovasc Surg. 2006; 32: 366-368Abstract Full Text Full Text PDF PubMed Scopus (35) Google Scholar2006Aortobronchial fistula00Ting6Ting A.C. Cheng S.W. Ho P. Chan Y.C. Poon J.T. Cheung G.C. Endovascular stent graft repair for infected thoracic aortic pseudoaneurysms—a durable option?.J Vasc Surg. 2006; 44: 701-705Abstract Full Text Full Text PDF PubMed Scopus (73) Google Scholar2006Mycotic aneurysms00Wheatley8Wheatley G.H. Nunez A. Preventza O. Ramaiah V.G. Rodriguez-Lopez J.A. Williams J. et al.Have we gone too far? Endovascular stent-graft repair of aortobronchial fistulas.J Thorac Cardiovasc Surg. 2007; 133: 1277-1285Abstract Full Text Full Text PDF PubMed Scopus (46) Google Scholar2007Aortobronchial fistula00Farber11Reisenmann P.J. Books J.D. Farber M.A. Thoracic endovascular repair of aortobronchial fistulas.J Vasc Surg. 2009; 50: 992-998Abstract Full Text Full Text PDF PubMed Scopus (38) Google Scholar2009Aortobronchial fistula02 (40%)Patel5Patel H.J. Williams D.W. Upchurch G.R. Dasika N.L. Eliason J.L. Deeb G.M. et al.Late outcomes of endovascular repair for the infected thoracic aorta.Ann Thorac Surg. 2009; 87: 1366-1371Abstract Full Text Full Text PDF PubMed Scopus (36) Google Scholar2009Mixed fistulae, mycotic aneurysms3 (15%)3 (15%) Open table in a new tab Our study was conducted to focus on outcomes following TEVAR for ITAP and represents one of the largest cohorts treated by an endovascular approach. The results suggest that this therapy can be performed safely with relatively low rates of early morbidity in a high-risk group. In this series, early mortality was 11.5%. There was no paraplegia or stroke in this cohort, and the rate of temporary dialysis was 7.4%. Late results remain poor, but the risk of late mortality is often related to the underlying comorbidities frequently present in this high-risk population. Risk of reinfection is significant, and our work is the first to suggest that the results are poorer in the setting of TEVAR for infected graft vs native aorta. While this finding is novel, the basis for this is supported by the reported (but low) rate of treatment success with antibiotic therapy alone. Antibiotic therapy may be able to sterilize native tissue but not infected graft, and wide debridement may be mandatory in that setting. Again, the parallel is drawn toward endocarditis where sterilization of valve infection without surgery is possible. In summary, thoracic endovascular repair is considered a suitable but palliative therapeutic option in patients presenting with infected thoracic aortic pathology. Late mortality in this high-risk group is often secondary to underlying pathology rather than to reinfection. Late reinfection may be more frequent when TEVAR is performed for prosthetic graft infection rather than native aortic infection. However, the low incidence of this disease makes a direct comparison to the gold standard of open repair difficult, and thus its role as either a definitive or bridge therapy remains elusive. Conception and design: HP, DW, GU, ND, JE, GDAnalysis and interpretation: HPData collection: HPWriting the article: HPCritical revision of the article: HP, DW, GU, ND, JE, GDFinal approval of the article: HP, DW, GU, ND, JE, GDStatistical analysis: HPObtained funding: HPOverall responsibility: HP
Referência(s)