| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Click here to read this article as a CME activity
Section of Thoracic Surgery, Department of Surgery, University of Michigan, Ann Arbor, Michigan
Accepted for publication March 6, 2009.
* Address correspondence to Dr Orringer, Section of Thoracic Surgery, University of Michigan Medical Center, 1500 East Medical Center Dr, 2120 Taubman Center, Box 0344, Ann Arbor, MI, 48109 (Email: morrin{at}umich.edu).
Presented at the Fifty-fifth Annual Meeting of the Southern Thoracic Surgical Association, Austin, TX, Nov 5–8, 2008.
| GENERAL 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 an individual non-member subscription to the journal.
|
 |
Abstract |
|---|
 Top Abstract IntroductionPatients and MethodsResultsCommentDiscussionFootnotesAcknowledgmentsReferences |
|---|
Methods: This single-institution retrospective study used medical records and an esophagectomy database to assess results in 1,133 patients who underwent transhiatal esophagectomy and a cervical esophagogastric anastomosis, 241 for benign disease and 892 for cancer, between January 1996 and December 2006.
Results: Esophagectomy patients who experienced CEGALs included 127 (14.2%) with cancer and 23 (9.5%) with benign disease. Logistic regression analysis identified increasing number of preoperative comorbidities (p < 0.001), active smoking history (p = 0.044), and postoperative arrhythmia (p = 0.002) as risk factors for CEGALs, and a side-to-side stapled cervical esophagogastric anastomosis compared with a manually sewn one as protective (p < 0.001). For cancer patients, higher pathologic stage disease (p = 0.050) was a risk factor for CEGALs. For patients with benign disease, a higher number of prior esophagogastric operations (p = 0.007) is a risk factor for CEGALs. Of the 90.7% of CEGALs that occurred on or before postoperative day 10, cervical wound drainage (63.3%) was the most common presenting symptom. Screening barium swallow identified postoperative complications and influenced outcome in 39 patients (3.8%).
Conclusions: Higher number of preoperative comorbidities, advanced pathologic stage, postoperative arrhythmia, an increased number of prior esophagogastric surgeries, and active smoking history are risk factors for developing CEGAL, and a side-to-side stapled cervical esophagogastric anastomosis is protective. Screening barium swallow identifies few postoperative complications, but provides quality control.
| The Hawley H. Seiler Resident Award is presented annually to the resident with the oral presentation and manuscript deemed the best of those submitted for the competition. This Award was inaugurated in 1997 to honor Dr Seiler for his contributions and dedicated service to the Southern Thoracic Surgical Association.
|
Esophagogastrectomy is the gold standard treatment for esophageal carcinoma and end-stage benign esophageal disease. The most common technique for esophageal resection at the University of Michigan is transhiatal esophagectomy (THE) with pyloromyotomy, feeding jejunostomy, and a cervical esophagogastric anastomosis (CEGA). One of the many advantages of a CEGA over an intrathoracic anastomosis is that a cervical esophagogastric anastomotic leak (CEGAL) is relatively easy to treat. The published rate of CEGAL is between 9% and 14% [1, 2], and a CEGAL may prolong hospital stay and lead to other complications [3].
The University of Michigan has a regimented postoperative pathway for patients undergoing THE with CEGA. On postoperative day (POD) 3, the nasogastric tube is removed, and the patient is started on sips of clear liquids. Oral intake is advanced in a stepwise fashion to clear liquids ad libitum on POD 4, full liquids on POD 5, and a "soft mushy" diet consisting of pureed meats and similar soft solid foods on POD 6. On POD 7, the patient undergoes a screening barium swallow (SBaS) to determine the presence of an anastomotic leak, gastric outlet obstruction at the level of the pylorus, decreased transit at the point of the feeding jejunostomy, or any other potential alimentary tract complication. The above THE pathway has contributed to an acceptable institutional mortality rate of 1% [2].
The University of Michigan experience with THE and a CEGA for a 10-year period was analyzed to define the uncomplicated THE postoperative pathway, identify patient risk factors for developing CEGALs and patterns of CEGAL presentation, and determine the value of a routine postoperative SBaS in detecting CEGALs and other complications.
|
Patients and Methods |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentDiscussionFootnotesAcknowledgmentsReferences |
|---|
The technique for THE has been described in detail [4]. The CEGA was performed either by a manually sewn technique or by a side-to-side stapled anastomosis with the Autosuture Endo GIA II 30-mm x 3.5-mm stapler (Covidien, Mansfield, MA) as previously described [5].
Description of Clinical Variables
Age, sex, and race were documented as preoperative characteristics. Race was defined as white and non-white (including Asian, black, Latino, and other). Smoking history was defined as never smoker, former smoker, and current smoker. Six preoperative comorbidities were measured: renal insufficiency (creatinine > 1.5 mg/dL), heart disease (history of previous myocardial infarction, coronary intervention or documented coronary artery disease, and diagnosis of congestive heart failure), diabetes (insulin and non–insulin-dependent), hypertension, previous stroke, and chronic obstructive pulmonary disease. Total morbidity was obtained by adding the number of the above comorbidities for each patient. For patients with benign esophageal disease, the primary diagnosis (including Barrett's with high-grade dysplasia, gastroesophageal reflux disease [GERD], achalasia, scleroderma, post–caustic injury, and other benign disease) and whether or not the THE was performed for perforation was defined, and the total number of prior esophagogastric operations, including fundoplications, myotomies, esophageal dilations, and other procedures, was determined for each patient. For patients with esophageal cancer, tumor histology (squamous cell carcinoma, adenocarcinoma, and other) and whether the patient had received preoperative chemotherapy or radiotherapy were recorded.
Intraoperative characteristics included mean intraoperative blood loss and whether the patient had a manually sewn or side-to-side stapled anastomosis, a partial sternotomy for exposure, a recurrent laryngeal nerve injury, or a thoracic duct injury. For cancer patients, the pathologic stage of tumor was noted. Postoperative characteristics before CEGAL for all patients included the occurrence of arrhythmias (atrial fibrillation or flutter, sinus tachycardia requiring medical treatment, bradycardia, ventricular ectopy requiring medical treatment, ventricular tachycardia) and respiratory insufficiency requiring intensive care unit admission or reintubation.
All CEGAL symptoms, drainage, fever greater than 100.5°F, erythema, swelling, crepitus, pain and halitosis, and their date of occurrence after THE were documented. A CEGAL was defined as a cervical anastomotic leak identified radiographically or clinically and confirmed after opening the neck wound and observing swallowed water issue from the wound.
Barium Esophagogram
Most patients underwent a screening barium swallow (SBaS) to evaluate the anastomosis, integrity of the conduit, and the pyloromyotomy site, and to rule out a small bowel obstruction at the site of the jejunostomy. Barium, not thinned, water-soluble contrast agents, is our contrast agent of choice for assessing anastomotic integrity.
Statistical Analysis
The goal of the analysis was to examine preoperative, intraoperative, and postoperative characteristics that are associated with CEGAL. We first assessed univariate associations between leak and covariates of interest. Pearson 
2 statistic and Fisher's exact test were used to assess the association between CEGAL (leak versus no leak) and categorical covariates such as sex, race, tumor histology (squamous cell carcinoma, adenocarcinoma, and other), anastomosis type (hand sewn versus side-to-side stapled), each of the above mentioned perioperative complications (yes or no), and diagnosis of benign subjects. Cochran-Armitage trend tests were used to assess the association between leak and an ordinal variable such as smoking history (never, former, and current smoker), total number of preoperative comorbidities, and total number of prior esophagogastric operations. Wilcoxon rank-sum tests were used to assess the association between leak and a continuous variable such as age and intraoperative blood loss.
For covariates that are considered clinically important to leak prognosis and covariates that are significantly correlated with leak in the univariate analysis, we then fit a logistic regression model to construct a multivariate model for leak likelihood. Three separate models were assessed, one for the benign cohort, one for the cancer cohort, and one for the benign and cancer cohorts combined. Patient characteristics and perioperative and postoperative variables were included in all three models. Variables unique to the benign model were total prior esophagogastric operations and benign disease diagnoses such as Barrett's with high-grade dysplasia, GERD, achalasia, and perforation. Variables unique to the cancer model were tumor histology, pathologic stage, preoperative chemotherapy, and preoperative radiation therapy. Backward deletion was used to select the variables for the final model. Variables that caused convergence difficulties because of complete separation and quasi-complete separation are taken out of the model and discussed separately.
All statistical analyses were done using SAS v9.1 (SAS Institute, Cary, NC). A two-tailed probability value of 0.05 or less was considered to be statistically significant.
|
Results |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentDiscussionFootnotesAcknowledgmentsReferences |
|---|
|
|
|
|
|
For patients undergoing THE for malignancy, higher pathologic stage (OR, 1.2; 95% CI, 1 to 1.5; p = 0.05), higher number of morbidity (OR, 1.4; 95% CI, 1.1 to 1.7; p = 0.001), and occurrence of arrhythmia (OR, 2.5; 95% CI, 1.3 to 4.6; p = 0.006) were associated with a higher probability of CEGAL (OR, 1.204; 95% CI, 1.000 to 1.454; p = 0.050), whereas a stapled anastomosis (OR, 0.4; 95% CI, 0.2 to 0.6; p = 0.0002) was associated with a lower probability of CEGAL.
Leak Characteristics
Cervical wound drainage, fever greater than 100.5°F, and erythema were the most prevalent symptoms in patients with CEGAL (63.3%, 25.3%, and 14.0%, respectively; Table 6). The earliest occurring symptoms were fever (mean POD 5.5 ± 2.9), followed by crepitus (mean POD 6.3 ± 4.6) and erythema (mean POD 6.6 ± 2.1). The exact day of leak could not be determined for 1 patient after review of the medical records, so an average POD for CEGAL was determined from the remaining 149 patients. The average POD for CEGAL was 7.4 ± 2.9 (range, 3 to 23 days), with no statistical difference in average day of CEGAL occurrence between patients undergoing THE for cancer or benign disease (p = 0.31). Only 9.3% of CEGAL patients experienced their anastomotic leaks after POD 10 (Fig 1). The 7-, 14-, and 21-day leak-free probabilities were 38.9%, 2.7%, and 0.7%, respectively (Fig 1).
|
|
Other Radiographic Screening Barium Swallow Findings Requiring Intervention
Of the 1,040 patients who underwent SBaS, 17 had radiographic findings other than CEGALs that required intervention. One patient had radiographic delayed pyloric emptying that required balloon dilation. Eleven patients had radiographic small bowel obstructions at the level of the feeding jejunostomy tube, resulting in 8 jejunostomy tube removals, 2 jejunostomy tube downsizings, and 1 exploratory laparotomy. Four patients had gastric conduit abnormalities: two gastric torsions, one a misinterpreted anastomotic leak that was called a gastric torsion and led to an operative intervention, and one gastric conduit leak. Also discovered was one diaphragmatic herniation of intestine alongside the gastric conduit requiring operative intervention.
|
Comment |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentDiscussionFootnotesAcknowledgmentsReferences |
|---|
Only a few reports have attempted to identify risk factors for the CEGAL. Potential intuitive preoperative risk factors for all patients would be preexisting diabetes, cardiovascular disease, and smoking history, as these patients, in theory, have reduced microperfusion of tissues. In support of this, in our study multivariate analysis identified positive smoking history as a risk factor for CEGAL. Dewar and colleagues [8] found no relationship between diabetes mellitus and cervical anastomotic leak. Our study found that diabetes mellitus, heart disease, chronic renal insufficiency, hypertension, previous stroke, and chronic obstructive pulmonary disease were not independent risk factors; however, in aggregate, the increasing number of comorbid conditions was a risk factor for experiencing CEGALs (mean total comorbidities 0.9 ± 1.1 versus 0.7 ± 0.9; OR, 1.423; 95% CI, 1.171 to 1.723; p = 0.001). This result reaffirms the intuitive notion that the sicker the patient preoperatively, the higher the likelihood of developing a CEGAL.
The diagnosis of cancer was a risk factor for a CEGAL as well as advanced pathologic stage. This result may be related to the clinical or subclinical state of malnourishment of these patients despite aggressive attempts to optimize nutrition preoperatively with enteric feedings. Patil and associates [9] found preoperative malnourishment as defined by an albumin concentration less than 3 g/dL to be a risk factor for CEGAL; however, markers of nutrition, such as albumin, prealbumin, and transferrin are not routinely measured in our patients preoperatively. Neoadjuvant chemoradiotherapy has been considered to be influential in leading to anastomotic leaks after esophagectomy. Briel and coworkers [10], evaluating cervical esophagogastric and esophagocolic anastomoses, found neoadjuvant therapy to be a statistically significant risk factor for anastomotic leak. However, most studies, including this one, do not confirm induction therapy as a risk factor for anastomotic leak [8, 9, 11, 12].
Relative ischemia of the tip of the newly mobilized gastric conduit has long been implicated in the higher rate of leak for the CEGA compared with an intrathoracic anastomosis. Because the short gastric and left gastric arteries are divided in the conduit mobilization, the tip of the conduit is supplied by collateral vessels from the right gastric and right gastroepiploic arteries [13]. As a result, careful technique, minimization of conduit trauma, and minimizing low-flow states are important in preventing anastomotic complications. Among our patients, those who experienced postoperative arrhythmias had a higher leak rate on univariate analysis (14.7% versus 5.4%; p < 0.001), and the occurrence of arrhythmia was predictive for leak on multivariate analysis. This observation may be a result of postoperative arrhythmia causing a low-flow state, putting an already mildly ischemic anastomosis at additional risk.
In the group with benign disease who underwent esophagectomy, patients who had a higher number of previous esophagogastric operations were predisposed to having CEGALs, suggesting that the increased operative trauma needed to take down the previously operated on gastric fundus may place the conduit at additional risk. Data for previous esophagogastric operations were prospectively collected from our database for patients undergoing an esophagectomy for benign disease. Prior esophagogastric operations included esophageal dilations for stricture, Botox injection and balloon dilation for achalasia, fundoplication for reflux, esophagomyotomy and fundoplication for achalasia, and primary repair of esophageal perforation, to name a few. It has been implied that a gastrostomy before esophagectomy may increase conduit complications by interfering with gastric blood supply and collateral circulation [14]. None of the benign leak patients had a previous percutaneous endoscopic gastrostomy tube or open gastrostomy tube, and in the benign nonleak group, 2 patients had a previous percutaneous endoscopic gastrostomy and 2 patients a previous open gastrostomy tube. The incidence of a prior gastrostomy tube is not enough to derive any sort of statistical correlation with subsequent anastomotic leak. With regard to prior gastrostomy tube placement in patients undergoing esophagectomy for malignancy, our group avoids them not only because of the potential for injury to the right gastroepiploic artery but also because of an increased risk of infection in a midline abdominal incision performed adjacent to a gastrostomy site. If enteral nutritional support is needed before esophagectomy, our preference is either a nasogastric Dobbhoff feeding tube, or in the rare event in which this is not possible, a jejunostomy tube. We estimate that fewer than 5% of our patients undergoing a THE have had a prior gastrostomy tube.
In contrast to the theory that transient changes in tissue perfusion, low-flow states, and reduction in oxygen carrying capacity may place the anastomosis at risk, we found no statistically significant differences in operative blood loss in our CEGAL versus non-CEGAL cohorts (mean blood loss (mL), 409 ± 356.5 versus 356.2 ± 320.0 mL; p = 0.158). However, measurement of intraoperative blood loss is a poor surrogate for relevant changes in blood flow and does not take into consideration perioperative hypotension and use of vasoactive agents that may place the anastomosis at risk. Intraoperatively, all patients have an arterial catheter placed for monitoring of blood pressure, especially during the transhiatal dissection in the posterior mediastinum when compression of the heart may occur. The use of vasoconstrictive agents such as norepinephrine and phenylephrine (Neo-Synephrine) to treat hypotension occurring during the transhiatal dissection is avoided by our anesthesiologists. Rather, the surgeon is kept informed if the blood pressure is too low, and the transhiatal dissection is intermittently interrupted until the patient is normotensive again. We rarely pass the conduit through the posterior mediastinum more than once, as we carefully orient the conduit before it is passed through the hiatus and verify its correct orientation by palpation once the tip is in the cervical wound. Postoperative hypotension in the euvolemic patient is often managed by adjusting the rate or composition of the epidural anesthetic agent or administering dopamine intravenously to enhance splanchnic perfusion. It is difficult to confirm whether alterations in gastric tip blood flow play a significant role in subsequent anastomotic leaks without some form of quantitative perioperative real-time measure of anastomotic tissue perfusion [15, 16], but identification of potential intraoperative factors that may affect conduit blood flow and anastomotic healing is a fertile area for future investigation.
The side-to-side stapled CEGA was developed in an attempt to reduce the rate of CEGALs [5]. The stapled anastomosis helped reduce the rate of CEGAL at the University of Michigan from 14% in patients before 1998 to 9% in patients from 1998 [2], with the majority of the anastomoses in the second group being stapled. This study confirms that a stapled anastomosis is protective against CEGALs. Other groups have published data supporting this finding as well [17, 18].
An interesting observation of this study is that patients undergoing THE for achalasia have less CEGALs than other THE patients (1.7%, 1 of 59 patients; p = 0.019). The cervical esophagus in advanced achalasia has a larger diameter, a thicker wall, and greater vascularity than normal, all factors that may contribute to a reduced CEGAL rate. It is difficult, however, to draw conclusions from these data given the small sample size (59 patients). Devaney and colleagues [19] evaluated THE in 93 patients operated on for end-stage achalasia. That study found an overall anastomotic leak rate of 10%; 4.8% (1 of 21) for patients who had a stapled CEGA, and 11.1% (8 of 72) in patients who had a hand-sewn CEGA. These leak results are similar to other published reports for CEGALs after THE for all diagnoses.
This study also sought to identify the most common postoperative symptoms of a CEGAL. Wound drainage (63.3%) was the most common symptom, followed by fever greater than 100.5°F (25.3%), erythema of the incision (14.0%), crepitus (12.0%), and swelling (10.7%). Unusual neck pain and halitosis were rare (6.7% and 2%, respectively). The earliest presenting symptoms were fever, crepitus, and erythema (mean POD onset of symptoms, 5.5 ± 2.9, 6.3 ± 4.6, 6.6 ± 2.1). Wound drainage was a relatively late-occurring symptom (mean 7.3 ± 2.7 days after THE). The majority of leaks occur on or before POD 10 (90.7%). However most patients are fit for discharge and tolerating solid food by POD 7 to 8 and would not benefit from an extended hospital stay simply in anticipation of a potential CEGAL. Therefore, patients are educated about the symptoms of a CEGAL, and those who live farther than 2 to 3 hours from the hospital are asked to stay locally for 3 to 7 days after discharge on POD 7 before returning home. Should symptoms of a leak present after discharge, the patients are instructed to return for evaluation and possible treatment.
On POD 7, a thin barium swallow or upper gastrointestinal series is obtained. The goal of the SBaS is to evaluate the anastomosis, the integrity of the conduit, and the patency of the pylorus as well as the caliber of the small bowel at the site of the jejunostomy. Barium is preferred in the screening test for anastomotic leak because it has a higher sensitivity for detecting a leak than thinner water-soluble contrast agents such as Gastrografin [20–22]. In addition, Gastrografin poses a risk of pneumonitis if aspirated [23]. The concept of barium-induced clinical mediastinitis is not supported in the literature [24]. Only one report has demonstrated histologic evidence of barium-induced mediastinitis experimentally in a feline animal model [25]. The animals demonstrated no additional clinical symptoms after barium was infused into their mediastinum compared with Gastrografin; although on necropsy, some granulomas were identified that were not seen in the Gastrografin-treated cohort. In addition, other groups have questioned the accuracy of water-soluble Gastrografin in detecting postesophagectomy anastomotic leaks [26].
This study found that 38.7% of CEGALs were detected clinically on the basis of symptoms and managed by opening the cervical wound empirically before an SBaS was obtained. Because of symptoms that developed after an indeterminate SBaS, 2.7% of patients with CEGALs had their cervical wounds opened. In addition, 44.7% of CEGALs in this series manifested after a negative SBaS had been obtained 1 week after THE. Whether those CEGALs were "missed" in a false-negative SBaS or developed after the SBaS is impossible to determine, but it is likely that the latter was the case in the majority of instances. In total, the SBaS obtained 1 week after the THE as a screening test led to the identification of 14.7% of the CEGALs that occurred in this series. Although this is a low yield for a screening tool, the SBaS also identified potentially catastrophic complications, such as gastric torsions, a nonanastomotic conduit leak, and a diaphragmatic hernia. Inasmuch as the SBaS is well tolerated by patients with virtually no complications of the test itself, the SBaS is an acceptable screening tool that provides quality control before a patient's discharge from the hospital after a THE. In addition, the SBaS can be used as a postoperative reference against which future barium swallows can be compared should the patient exhibit dysphagia in follow-up as an outpatient.
In conclusion, this is the largest reported series to identify risk factors for CEGALs, characterize the most common presenting symptoms, and assess the merits of the screening contrast barium swallow examination obtained 1 week after THE. These data should serve as a reference guide for the clinician performing THE and a CEGA and potentially assist in the prevention and detection of CEGALs.
|
Discussion |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentDiscussionFootnotesAcknowledgmentsReferences |
|---|
You stated in the manuscript that one of the purposes of your retrospective study was to evaluate the postoperative pathway of your patients. You note that barium swallow was performed on day 7 and it identified only 3.8% of the possible situations that led to a change in your therapy, and that was 15% of the 15% of patients that had leak, more or less. So you do the math there, I don't know what that was, but it is low. So despite the low diagnostic yield, you continue to recommend barium swallow. My question is this. Would it make sense perhaps to perform the barium swallow prior to instituting PO (per os) intake, which you do on postoperative day 5, I believe, rather than waiting until postoperative day 7, in which case you could diagnose things such as silent aspiration, gastric outlet obstruction, J-tube obstruction, et cetera, prior to instituting feeding? And in that way you can continue to rely on the clinical factors that you already rely on for cervical esophagogastric leak.
The second question is that you noted that atrial fibrillation is associated with postoperative leak. You theorize in your comments that the leak may be due to a low-flow state induced by atrial arrhythmia. Is it possible that the esophagogastric leak causes excitability of the heart and that the atrial fibrillation is a manifestation rather than a contributor to the leak? We have also noticed the same finding in intrathoracic leaks, that the incidence of atrial fibrillation is higher in patients with arrhythmia.
Our service has also converted to using CT (computed tomographic) scans with PO contrast to diagnose the presence or extent of a suspected leak. That has gone a long way in telling us what to do next. In other words, is this a contained leak or is it a noncontained leak and how clinically significant is it? So really when we suspect leak, we are going over to a CT scan for that. Is there any role for CT scan in cervical leak detection/management?
Finally, and I know I am asking a lot of questions but this one is rhetorical, we recently presented a paper comparing different methods of anastomosis for the intrathoracic anastomosis as opposed to the transhiatal, which you are presenting here. Interestingly, the side-to-side stapled anastomosis showed a 9% leak rate in our series as well—and this is intrathoracic again—compared to 4% for a circular stapler. Do you foresee the cervical leak rate dropping below 9%, or are you ready to holler "uncle," abandon the cervical anastomosis, and adopt the higher-performance circular intrathoracic anastomosis?
DR COOKE: Thank you, Dr Hofstetter, for those interesting comments. I am glad to present the 151st paper out of the University of Michigan with regards to cervical anastomosis.
In regards to your first question, should we do the barium swallow earlier, we perform the barium swallow on day 7. That allows us to do a stepwise progression of oral intake. Oftentimes the patient needs a little bit of time to recover from the surgery. The barium swallow is a test that requires the patient to actively swallow, in a relatively aggressive manner, a large amount of barium, and on postop day 3 where there is pain in the neck, et cetera, and the patient is just getting ready to mobilize more frequently out of bed, the yield, or the completion of that study, may not be as effective in the early stages as opposed to the later stages.
DR HOFSTETTER: So along those same lines, why are you in a hurry to feed them? Why not wait until they are ready to take a barium swallow before you feed them?
DR COOKE: That, again, goes into the global picture of the advantages of a cervical anastomosis. If there is a leak with a cervical anastomosis then the wound can be opened and packed, and the patient can still drink and eat despite that packing as long as they are able to manage their own wound. The fistula closes within 2 weeks, the patients will go home with a fistula, and it closes very simply.
In regards to your next question, arrhythmia, is it the chicken or the egg? That we can't really tell by the nature of the study, and it is difficult for us to tell unless we get a barium swallow or a CT scan at the time of the onset of arrhythmia. What I didn't talk about in the slides but I do talk about in the manuscript is that arrhythmia is a broad category that includes atrial fibrillation, symptomatic bradycardia, symptomatic sinus tachycardia requiring medical intervention, in a few patients who have had actual ventricular abnormalities requiring medical intervention. So it is a broad category. So in that respect, many of those things aren't necessary, such as bradycardia is not necessarily a symptom of a leak per se, but could they be low-flow states in an already possibly ischemic anastomosis? But again, your point is valid, is this the chicken or the egg?
In regards to CT-guided detection of anastomotic leak, I am not familiar with the use of CT for those purposes, and I don't know the cost-benefit analysis of CT versus barium swallow, which is a very cheap and inexpensive test.
And finally, in response to your last question in regards to why I don't just do an intrathoracic anastomosis, I believe in a cervical anastomosis mainly because of what I discussed before, and that is the ease of complication. A cervical anastomosis is easy to deal with. That being said, there can be instances where an Ivor Lewis esophagectomy might be the appropriate procedure for that patient. And I think it is important to look at the patient from an individual standpoint and see what the best is, but I do bend over backwards to do a cervical anastomosis, because the complications are easier to deal with.
DR RICHARD H. FEINS (Chapel Hill, NC): During the last 20 years or so of doing this procedure, there was a time when we fell into the practice of really making a nice, thin tube, and what we found is we could bring it right up to the mediastinum very easily and had a lot of room, particularly as we switched over to a stapled anastomosis. The problem with the thin tube was it leaked roughly 100% of the time. And in talking with Jim Luketich about it, and he has a paper to this effect, he mentioned that the minimum that that tube should be is about 6 cm, and we switched back and since that time have had very little, if any, leaking. So I am wondering what your policy is about the width of your conduit and also whether or not the degree of cancer and the degree of resection you have to do, which did result in higher leaks, might in part be related to the width of your conduit?
DR COOKE: We do tubularize our conduit with a GI (gastrointestinal) stapler. Whether that leads to further damage to the conduit itself and to the predisposition of leak, I am not sure, but we routinely do that. It does help with the gastric emptying of our conduit to tubularize it in adding on to a pyloric myotomy.
DR FEINS: How wide is it?
DR COOKE: It depends on the level of the tumor itself, but generally it is about 4 to 5 cm in width. But whether that predisposes to an anastomotic leak, I am not sure.
|
Acknowledgments |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentDiscussionFootnotesAcknowledgmentsReferences |
|---|
|
Footnotes |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentDiscussionFootnotesAcknowledgmentsReferences |
|---|
|
References |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentDiscussionFootnotesAcknowledgmentsReferences |
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |
