Portal de Periódicos da CAPES

Bleeding Risk of Transesophageal Echocardiography in Patients With Esophageal Varices

2019; Elsevier BV; Volume: 32; Issue: 5 Linguagem: Inglês

10.1016/j.echo.2018.11.017

1097-6795

Abiy Nigatu, John Erikson Yap, Katherine Lee Chuy, Benjamin Go, Rami Doukky,

Abdominal vascular conditions and treatments

Gastrointestinal (GI) bleeding is an important complication of transesophageal echocardiography (TEE) in patients with esophageal varices complicating advanced liver disease, as variceal bleeding can be fatal given coexisting coagulopathy.1Daniel W.G. Erbel R. Kasper W. Visser C.A. Engberding R. Sutherland G.R. et al.Safety of transesophageal echocardiography. A multicenter survey of 10,419 examinations.Circulation. 1991; 83: 817-821Crossref PubMed Scopus (726) Google Scholar, 2Hahn R.T. Abraham T. Adams M.S. Bruce C.J. Glas K.E. Lang R.M. et al.Guidelines for performing a comprehensive transesophageal echocardiographic examination: recommendations from the American Society of Echocardiography and the Society of Cardiovascular Anesthesiologists.J Am Soc Echocardiogr. 2013; 26: 921-964Abstract Full Text Full Text PDF PubMed Scopus (711) Google Scholar The 1-year risk of spontaneous bleeding from esophageal varices ranges between 1.6% and 68.9%, based on morphologic and clinical parameters.3North Italian Endoscopic Club for the Study and Treatment of Esophageal VaricesPrediction of the first variceal hemorrhage in patients with cirrhosis of the liver and esophageal varices. A prospective multicenter study.N Engl J Med. 1988; 319: 983-989Crossref PubMed Scopus (1074) Google Scholar The blinded instrumentation of the esophagus with a TEE probe may increase this risk. We investigated the rate of TEE-related GI bleeding in patients with esophageal varices at our institution and pooled our data with a systematic review of published studies addressing the same question. In a retrospective cross-sectional study design, we queried our institutional electronic structured reporting database for all TEE procedures performed in the period from January 1, 2006, to December 31, 2016 (n = 2,208). Subjects who underwent an esophagogastroduodenoscopy (EGD) within 1 year before or after the TEE were identified (n = 301). EGD reports were reviewed to identify subjects with EGD evidence of esophageal varices (n = 20; Supplemental Figure 1, available at www.onlinejase.com). The electronic health records of the 20 subjects who met the inclusion criteria were reviewed to determine demographic, clinical, and laboratory data, as well as outcome events of TEE-related procedural or postprocedural bleeding. Bleeding events were defined as hematemesis, melena, hematochezia, packed red blood cell transfusion, or hemoglobin decrease by ≥2 g/dL from baseline within 48 hours after the procedure. The baseline characteristics of the study subjects are summarized in Table 1. All subjects underwent in-patient TEE and remained hospitalized for ≥48 hours after the TEE. The mean model for end-stage liver disease (MELD) score was 17.8 ± 8.5, the mean platelet count was 90 ± 52 × 103/μL, and the mean international normalized ratio (INR) was 1.44 ± 0.35. Most (80%) of patients had small (grade 1) esophageal varices. None of the study subjects had a documented overt GI bleeding or had a significant decrease in hemoglobin level.Table 1Baseline characteristicsCharacteristicValueTotal no. of patients20Age, yr57 ± 11Male17 (85)Race Hispanic11 (55) Caucasian4 (20) African-American1 (5) Others4 (20)Indication for TEE Infective endocarditis13 (65) Valvular heart disease3 (15) Thromboembolic disease/stroke3 (15) Shunt evaluation1 (5)Etiology of cirrhosis Alcoholic liver disease13 (65) Viral hepatitis C4 (20) Unknown3 (15)History of upper GI bleeding7 (35)TEE performed prior to diagnosis of varices6 (30)Beta-blocker use5 (25)Prior endoscopic variceal treatment4 (20)Child-Pugh class A3 (15) B13 (65) C4 (20)MELD score 206 (30)Esophageal varices size Grade 116 (80) Grade 2 or 34 (20)Hemoglobin, g/dL10.1 ± 1.85INR1.44 ± 0.35Platelet count, 103/μL90 ± 52Continuous data were presented as means ± standard deviations, and dichotomous variables were presented as frequencies (%). Open table in a new tab Continuous data were presented as means ± standard deviations, and dichotomous variables were presented as frequencies (%). After analyzing our institutional data, we performed a systematic literature search in PubMed and EMBASE to identify published original investigations reporting TEE-related bleeding in the period from January 1, 1971, to December 31, 2017 (Supplemental Figure 2, available at www.onlinejase.com). The search terms yielded 955 articles, of which seven reported a TEE-related bleeding complication. Including our own series, a total of eight studies, encompassing 1,078 subjects, were ultimately analyzed. The primary outcome was the composite of overt upper GI bleeding (hematemesis or melena), transfusion of packed red blood cell, or a decrease in hemoglobin level by ≥2 g/dL below baseline. Table 2 summarizes eligible studies and the baseline characteristics of the enrolled subjects. All eight studies were single-center case series, out of which five investigated intraoperative TEE in patients undergoing orthotopic liver transplantation,4Suriani R.J. Cutrone A. Feierman D. Konstadt S. Intraoperative transesophageal echocardiography during liver transplantation.J Cardiothorac Vasc Anesth. 1996; 10: 699-707Abstract Full Text PDF PubMed Scopus (86) Google Scholar, 5Burger-Klepp U. Karatosic R. Thum M. Schwarzer R. Fuhrmann V. Hetz H. et al.Transesophageal echocardiography during orthotopic liver transplantation in patients with esophagoastric varices.Transplantation. 2012; 94: 192-196Crossref PubMed Scopus (55) Google Scholar, 6Myo Bui C.C. Worapot A. Xia W. Delgado L. Steadman R.H. Busuttil R.W. et al.Gastroesophageal and hemorrhagic complications associated with intraoperative transesophageal echocardiography in patients with model for end-stage liver disease score 25 or higher.J Cardiothorac Vasc Anesth. 2015; 29: 594-597Abstract Full Text Full Text PDF PubMed Scopus (27) Google Scholar, 7Pai S.L. Aniskevich 3rd, S. Feinglass N.G. Ladlie B.L. Crawford C.C. Peiris P. et al.Complications related to intraoperative transesophageal echocardiography in liver transplantation.Springerplus. 2015; 4: 480Crossref PubMed Scopus (21) Google Scholar, 8Markin N.W. Sharma A. Grant W. Shillcutt S.K. The safety of transesophageal echocardiography in patients undergoing orthotopic liver transplantation.J Cardiothoracic Vasc Anesth. 2015; 29: 588-593Abstract Full Text Full Text PDF PubMed Scopus (39) Google Scholar whereas the remaining three series (including ours) were in general clinical settings.9Spier B.J. Larue S.J. Teelin T.C. Leff J.A. Swize L.R. Borkan S.H. et al.Review of complications in a series of patients with known gastro-esophageal varices undergoing transesophageal echocardiography.J Am Soc Echocardiogr. 2009; 22: 396-400Abstract Full Text Full Text PDF PubMed Scopus (57) Google Scholar, 10Pantham G. Waghray N. Einstadter D. Finkelhor R.S. Mullen K.D. Bleeding risk in patients with esophageal varices undergoing transesophageal echocardiography.Echocardiography. 2013; 30: 1152-1155Crossref PubMed Scopus (20) Google Scholar The predominant indication for TEE was hemodynamic monitoring during surgery (90%), followed by investigation for suspected infective endocarditis (7%). The weighted mean MELD score, a surrogate of liver disease severity, was 25 ± 8, with a mean weighted platelet count of 80 × 103/μL and mean INR of 1.9. Fifteen (1.39%) patients (95% confidence interval, 1.35%-1.43%) had a decrease in hemoglobin level or an overt GI bleeding after TEE. The study by Myo Bui et al.6Myo Bui C.C. Worapot A. Xia W. Delgado L. Steadman R.H. Busuttil R.W. et al.Gastroesophageal and hemorrhagic complications associated with intraoperative transesophageal echocardiography in patients with model for end-stage liver disease score 25 or higher.J Cardiothorac Vasc Anesth. 2015; 29: 594-597Abstract Full Text Full Text PDF PubMed Scopus (27) Google Scholar reported a high bleeding rate (2.54%) but notably used a MELD score ≥25 as an entry criterion and considered a bloody nasogastric tube output of >500 mL as a GI bleed. After excluding the latter study, given its unusual methods, the recalculated bleeding risk was only 0.62% (95% confidence interval, 0.61%-0.62%). The pooled bleeding risk in the five studies in which TEEs were performed intraoperatively was 1.4% (a third of which were overt GI bleedings), while the pooled bleeding risk in the three studies in which TEEs were performed in general clinical settings was 0%.Table 2Summary of studies included in systematic reviewAuthorsYearSettingNBleeding incidence, %Large varices, %Platelet count, 103/μLINRMELD scoreSuriani et al.4Suriani R.J. Cutrone A. Feierman D. Konstadt S. Intraoperative transesophageal echocardiography during liver transplantation.J Cardiothorac Vasc Anesth. 1996; 10: 699-707Abstract Full Text PDF PubMed Scopus (86) Google Scholar1996Intraoperative234.35NANANANASpier et al.9Spier B.J. Larue S.J. Teelin T.C. Leff J.A. Swize L.R. Borkan S.H. et al.Review of complications in a series of patients with known gastro-esophageal varices undergoing transesophageal echocardiography.J Am Soc Echocardiogr. 2009; 22: 396-400Abstract Full Text Full Text PDF PubMed Scopus (57) Google Scholar2009Consult- based140421371.618Burger-Kelpp et al.5Burger-Klepp U. Karatosic R. Thum M. Schwarzer R. Fuhrmann V. Hetz H. et al.Transesophageal echocardiography during orthotopic liver transplantation in patients with esophagoastric varices.Transplantation. 2012; 94: 192-196Crossref PubMed Scopus (55) Google Scholar2012Intraoperative2870.355478NA (PT = 51 sec)16Pantham et al.10Pantham G. Waghray N. Einstadter D. Finkelhor R.S. Mullen K.D. Bleeding risk in patients with esophageal varices undergoing transesophageal echocardiography.Echocardiography. 2013; 30: 1152-1155Crossref PubMed Scopus (20) Google Scholar2013Consult-based24037142NA15Myo Bui et al.6Myo Bui C.C. Worapot A. Xia W. Delgado L. Steadman R.H. Busuttil R.W. et al.Gastroesophageal and hemorrhagic complications associated with intraoperative transesophageal echocardiography in patients with model for end-stage liver disease score 25 or higher.J Cardiothorac Vasc Anesth. 2015; 29: 594-597Abstract Full Text Full Text PDF PubMed Scopus (27) Google Scholar2015Intraoperative4332.54NA661.932Pai et al.7Pai S.L. Aniskevich 3rd, S. Feinglass N.G. Ladlie B.L. Crawford C.C. Peiris P. et al.Complications related to intraoperative transesophageal echocardiography in liver transplantation.Springerplus. 2015; 4: 480Crossref PubMed Scopus (21) Google Scholar2015Intraoperative1610.6229931.9NAMarkin et al.8Markin N.W. Sharma A. Grant W. Shillcutt S.K. The safety of transesophageal echocardiography in patients undergoing orthotopic liver transplantation.J Cardiothoracic Vasc Anesth. 2015; 29: 588-593Abstract Full Text Full Text PDF PubMed Scopus (39) Google Scholar2015Intraoperative1160.86NA1042.125Nigatu et al.∗The present case series.2018Consult-based20015901.418Pooled data1,0781.39%†Pooled incidence.43†Pooled incidence.80‡Weighted means.1.9‡Weighted means.25‡Weighted means.NA, Not available; PT, Prothrombin time.The study design for all studies was a case series.∗ The present case series.† Pooled incidence.‡ Weighted means. Open table in a new tab NA, Not available; PT, Prothrombin time. The study design for all studies was a case series. Our case series show that patients with esophageal varices who undergo a TEE have a low incidence of bleeding. Our systematic literature review yielded a small number of case series, which affirmed the low risk of TEE-related bleeding in patients with esophageal varices. Bleeding risk was highest (1.4%) in the context of intraoperative TEEs performed in patients undergoing liver transplantation, while the risk was very low (approaching 0%) in the general clinical setting. The absence of bleeding in three case series from general clinical settings may be explained by the small sample size (58 subjects), which may be insufficient to uncover uncommon events. Moreover, selection bias may have contributed to lack of bleeding in these cohorts, as cardiologists may have chosen not to perform TEE on patients deemed to be at high bleeding risk. In addition, it is plausible that these TEEs are of shorter duration (in-and-out) and possibly limited to midesophageal views. On the other hand, patients who had intraoperative TEEs probably had more advanced liver disease, were at higher bleeding risk, and likely had a prolonged esophageal intubation. Notably, the study by Myo Bui et al.,6Myo Bui C.C. Worapot A. Xia W. Delgado L. Steadman R.H. Busuttil R.W. et al.Gastroesophageal and hemorrhagic complications associated with intraoperative transesophageal echocardiography in patients with model for end-stage liver disease score 25 or higher.J Cardiothorac Vasc Anesth. 2015; 29: 594-597Abstract Full Text Full Text PDF PubMed Scopus (27) Google Scholar which reported a relatively high bleeding rate (2.54%), investigated patients with the highest MELD scores, the lowest platelet counts, and the highest INRs, suggesting that advanced liver disease is a risk factor for TEE-related variceal bleeding. Clearly, our and other case series are limited by their observational noncontrolled design and selection bias. In conclusion, in patients with esophageal varices undergoing TEE, the incidence of TEE-related variceal bleeding is low, particularly among patients with small varices undergoing TEE in general clinical settings. These findings challenge the current practice of considering the presence of esophageal varices a relative contraindication for TEE. Large controlled studies or clinical registry data, such as ImageGuideEcho,11Tilkemeier P.L. Doukky R. Kirkpatrick J.N. Desai M.Y. Nagueh S.F. Coming-of-age: the ImageGuide Registry at three.J Nucl Cardiol. 2018; https://doi.org/10.1007/s12350-018-1442-1Crossref Scopus (4) Google Scholar are needed to investigate the risk of bleeding among patients with large esophageal varices and explore the significance of severity of liver disease in determining bleeding risk. Supplemental Figure 2Selection of studies for the systematic review. Search Terms: transesophageal echocardiogram + complications; bleeding/hematemesis/melena + esophageal varices; transesophageal echocardiogram + bleeding; transesophageal echocardiogram + esophageal varices.View Large Image Figure ViewerDownload Hi-res image Download (PPT)