Ann Thorac Surg 2011;92:2041-2045. doi:10.1016/j.athoracsur.2011.08.013
© 2011 The Society of Thoracic Surgeons
Original Articles: General Thoracic
Hiatal Hernia After Esophagectomy: Analysis of 2,182 Esophagectomies From a Single Institution
Theolyn N. Price, MD,
Mark S. Allen, MD*,
Francis C. Nichols, III, MD,
Stephen D. Cassivi, MD, MS,
Dennis A. Wigle, MD, PhD,
K. Robert Shen, MD,
Claude Deschamps, MD
Division of General Thoracic Surgery, Mayo Clinic, Rochester, Minnesota
Accepted for publication August 2, 2011.
* Address correspondence to Dr Allen, Division of General Thoracic Surgery, Mayo Clinic, 200 First St SW, Rochester, MN 55905 (Email: allen.mark{at}mayo.edu).
Presented at the Poster Session of the Forty-seventh Annual Meeting of The Society of Thoracic Surgeons, San Diego, CA, Jan 31–Feb 2, 2011.
 |
Abstract
|
|---|
Background: Esophageal resection is a complex operation often associated with morbidity. Hiatal hernia after esophagectomy is an unusual complication. We reviewed our experience with this complication.
Methods: From February 1988 through February 2009 we performed 2,182 esophagectomies. Fifteen (0.69%) patients experienced a hiatal hernia. We reviewed our prospective database for demographics, presentation, operative approaches, and outcomes.
Results: There were 14 men and 1 woman with a mean age of 59 years. Hernia developed after Ivor Lewis approach in 9, transhiatal in 5, and substernal colon interposition in 1. Presenting symptoms included pain in 7 patients, obstructive symptoms in 5, high chest tube output in 2, shortness of breath in 2, diarrhea in 1, and cough with dysphagia in 1. Two patients were asymptomatic. Radiographic studies revealed bowel in the left chest in 11 patients, right chest in 2, bilaterally in 1, and posterior mediastinum in 1. Hernia repair was through the abdomen in 14 patients and left chest in 1. All had reduction of the herniated contents and closure of the defect; 2 required mesh. There was no early mortality. Complications included wound infection, deep venous thrombosis, chylothorax, urinary retention, sacral decubiti, atrial arrhythmias, respiratory failure, and empyema. Mean follow-up was 34 months. Ten patients are still alive. There have been two hernia recurrences.
Conclusions: Hiatal hernia after esophagectomy is rare. Repair can be accomplished with low mortality; however, there is substantial morbidity. Because of the increased risk of incarceration or strangulation, these herniae should be repaired. Long-term outcome is usually excellent.
|
Introduction
|
|---|
Esophagectomy is a complex operation with the potential for substantial morbidity. When resecting the esophagus, the diaphragmatic hiatus often requires widening to allow the conduit to pass freely into the chest and prevent conduit obstruction. However, enlargement of the hiatus increases the risk of a hernia developing and allowing abdominal contents to pass into the chest. We reviewed our experience with esophagectomy to estimate the incidence of this rare type of hernia, and to describe the presentation and management of this problem.
|
Patients and Methods
|
|---|
From February 1988 through February 2009 we performed 2,182 esophageal resections at Mayo Clinic in Rochester, Minnesota. A retrospective review of these patients was performed of our prospectively maintained general thoracic surgery database. There were 15 (0.69%) patients who required reoperation at our institution to repair herniation of abdominal contents through the diaphragmatic hiatus after esophagectomy. Medical records were reviewed for demographics, presentations, operative procedures (approaches for primary resection and herniorrhaphy), and outcomes. Morbidity was defined as any postoperative complication occurring within 30 days of the index procedure or during that hospitalization. Mortality was defined as death within 30 days of the index procedure or during that hospitalization. Baseline characteristics and event rates were reported as a median with the data ranges. Follow-up for patients who had herniorrhaphy was complete in 100%, and was defined as the time between the herniorrhaphy and the last known follow-up. Survival was reported as the patient's status at the time of last follow-up. The Mayo Foundation Institutional Review Board approved this study.
|
Results
|
|---|
There were 14 men and 1 woman; the median age when herniation was discovered was 61 years (range, 34 to 76 years). All initial esophagectomies, with the exception of one, were performed at Mayo Clinic, Rochester, Minnesota. Seventy-two percent of all of the esophageal resections performed during this period were performed through an Ivor Lewis or transhiatal approach. Of the 15 patients who experienced diaphragmatic hernias, an Ivor Lewis approach was used in 9 patients, and a transhiatal approach was used in 5. One patient required reconstruction with an esophagojejunal free graft combined with a substernal right colon interposition. There was no difference in the hernia incidence when comparing an Ivor Lewis approach (9 of 978; 0.92%) and a transhiatal approach (5 of 601; 0.83%). Histopathologic examination indicated adenocarcinoma in 13 patients, squamous cell carcinoma in 1, and high-grade dysplasia in 1. Preoperative radiation and chemotherapy was administered to 9 patients. One patient received preoperative radiation only. All of the cancers were located in the distal esophagus. The pathologic TNM stage was 0 in 2 patients, py0 in 5, IB in 1, pyIIB in 1, IIIA in 1, pyIIIA in 4, and stage IV in1. Thirteen patients underwent widening of the esophageal hiatus at the time of the original esophagogastrectomy.
Presenting symptoms of herniation included pain in 7 patients, nausea and vomiting with obstructive symptoms in 5, high chest tube output in 2, shortness of breath in 2, diarrhea in 1, and cough and dysphagia with a tracheoesophageal fistula in 1. Two patients were asymptomatic, with hernias found incidentally on follow-up imaging. All patients had radiographic evidence of herniation; either a chest roentgenogram or computed tomography demonstrated loops of bowel in the left chest in 11 and in the right chest in 2. One patient had small bowel in bilateral pleural spaces, and another had small bowel contained in the posterior mediastinum. Median time between the original esophagectomy and repair of the hernia was 1 year 9 months (range, 3 days to 12 years). An elective operation was performed in 13 patients, and 2 required emergent procedures.
At reoperation, colon was found within the hernia sac in 6 patients, small bowel in 5, and both in 4. Ischemic small bowel and colon were found in 1 patient each, requiring partial small bowel resection and subtotal colectomy, respectively. One patient required partial omentectomy because of necrotic omentum. Herniorrhaphy was through the abdomen in 14 patients and left chest in 1. All had reduction of the herniated contents and closure of the defect around the conduit. Mesh was required to repair the defect in 2 patients. There were no injuries to the conduit. Median length of hospitalization after reoperation was 10 days (range, 4 to 28 days). For those without complications, the median length of hospitalization was 7 days (range, 4 to 10 days), but for those with complications, the median length of hospitalization was 15 days (range, 5 to 28 days). Complications occurred in 9 of the 15 patients (60%) and included wound infection, deep venous thrombosis, atrial arrhythmias, anemia, chylothorax requiring reoperation, urinary retention, sacral decubitus ulceration, pneumonia, respiratory failure, and empyema requiring decortication. There was no operative mortality.
Follow-up was complete in all patients who underwent a herniorrhaphy for a median of 1 year 11 months (range, 2 months to 14 years). There have been 5 deaths occurring at 5, 9, 12, 28, and 69 months after repair of the hernia. Three patients died of their esophageal malignancy, 2 patients died of unknown cause, 8 are alive with no evidence of disease, and 2 are alive with recurrence of their esophageal malignancy. There have been two hernia recurrences (13.3%). One occurred during the hospitalization after initial repair before discharge, but because it was small (1 to 2 cm) and completely asymptomatic, it was not initially re-repaired. However, re-repair was required 26 months later because of shortness of breath secondary to a large amount of bowel in the left chest. The second recurrence occurred 8 months after the initial repair, and was repaired through a laparotomy while a simultaneous ventral hernia repair was performed. At 4 months' follow-up, he has not had further recurrence.
|
Comment
|
|---|
Diaphragmatic hernia after an esophagectomy is an unusual complication that we found to have an incidence of slightly greater than 1 in 200 patients. We did not specifically look for a diaphragmatic hernia in all patients after an esophagectomy and many have continued follow-up at another institution; therefore, some may have experienced a hernia we are not aware of, so the overall incidence may be higher. There are 82 reported patients with hiatal hernia after esophagectomy in the English literature [1–26]. The incidence of this complication ranges from 0.4% to 6% after open esophagectomy [1]. After a minimally invasive esophagectomy, the incidence appears to be slightly higher at 2.7% to 4.5% [1, 2, 14, 19]. Kent and colleagues reported the incidence of diaphragmatic hernia after esophagectomy to be 0.8% with the open approach and 2.8% with the minimally invasive esophagectomy [1]. The hernia can occur as an early or late complication. In the early postoperative period, the most likely explanation is that there are fewer peritoneal adhesions [2]. With longer follow-up, it is theorized this complication is a result of progressive hiatal dilation as a result of increased intraabdominal pressure and suction effect of the negative intrathoracic pressure. This result is magnified by the increased use of minimally invasive approaches that decrease the amount of postoperative adhesions [2, 13, 14]. An alternative explanation regarding minimally invasive esophagectomy may be that when performing the esophageal mobilization, the hiatus is made larger than with the open technique because of the distortion of the hiatus from the abdominal insufflation or the need to have a larger passage way to avoid undue tension on the conduit as it is pulled up to the neck or upper thorax. Another possible explanation for the low reported incidence of this postoperative complication is the overall poor survival after esophagectomy for malignant disease. In patients with locally advanced disease, long-term cure rates of 25% to 30% are commonly reported. It is worth noting that of the 15 patients in our series, 9 had very early stage disease.
Presentation of patients with a diaphragmatic hernia after an esophagectomy is variable. The scope ranges from a small portion of patients who present with no specific symptoms to others who present with a myriad of symptoms. Some of the various reported symptoms include respiratory distress, intestinal obstruction, pain (chest or abdomen), lower gastrointestinal bleeding, fever, leukocytosis, jaundice, and liver congestion [10, 12, 14, 19]. Chest roentgenograms will usually show an increased density or air fluid levels in the lower portion of the left pleural space, but can also show retrocardiac air or even an increased right-sided density or air fluid levels (Fig 1) [6, 11, 14, 16, 21]. Computed tomography is diagnostic as it clearly shows herniated abdominal contents in the thoracic cavity (Figs 2, 3)
. Contrast swallow studies are usually unnecessary. Occasionally, asymptomatic patients are discovered to have a hernia on follow-up scans obtained as part of standard oncologic evaluation. Even if patients do not have symptoms, the hernia is likely to enlarge with time, increasing the chance of incarceration or strangulation, making the repair more complicated. This substantially increases the risk of morbidity and mortality. Based on these facts, it is our opinion that these diaphragmatic hernias should be surgically repaired at the time of diagnosis. When reviewing the literature, most authors concur, and it is the overwhelming opinion that surgical repair should be strongly considered in these patients. Repair should only be deferred for those with very small asymptomatic hernias and those with significant comorbidities or a short life expectancy owing to known malignancy [1–3, 12, 13, 19].
View larger version (116K):
[in this window]
[in a new window]
|
Fig 1. Chest radiograph demonstrating herniated colon into the left chest after an Ivor Lewis esophagogastrectomy.
|
|
View larger version (112K):
[in this window]
[in a new window]
|
Fig 2. Computed tomography demonstrating herniated colon into the left chest after an Ivor Lewis esophagogastrectomy.
|
|
View larger version (114K):
[in this window]
[in a new window]
|
Fig 3. Computed tomography demonstrating herniated colon into the right chest after an Ivor Lewis esophagogastrectomy.
|
|
Repair of the hernia is usually straightforward. We used a transabdominal approach in the majority of these patients, which allows easy reduction of the hernia sac contents. If there is ischemic or necrotic tissue, it should be resected and reconstructed as necessary. The hiatus should be closed snugly enough around the gastric conduit to minimize the chance of recurrent hernia, but keeping in mind not to place undue tension on the gastric conduit, possibly jeopardizing its vascular supply. This is accomplished by placing nonabsorbable suture into the edges of the crura, making the hiatus smaller. Kent and colleagues stress the importance of completely mobilizing the crura posterior to the conduit to allow for a tension-free closure. Extreme care must be taken not to injure the vascularity of the conduit (the right gastroepiploic artery). They also recommend preservation of the peritoneal lining of the crura to add integrity to the repair [1]. In our series, this approach was always successful in reducing the herniated viscera and allowing repair of the hiatus. In situations in which the tension is too great for primary repair of the hiatus, one could consider a relaxing incision in the left hemidiaphragm or the use of synthetic material, as was done in 2 of our patients. Some advocate the use of mesh to enhance the strength of hernia repair, but others would argue that mesh should not be used in this location because it is not usually necessary and it may potentially erode into the conduit, its blood supply, or other adjacent structures [1, 6, 10, 12, 14]. If laparotomy is not an option because of excessive adhesions or some other contraindication, then a left thoracotomy should be considered. This was done in 1 of our patients to avoid the mesh from a previously repaired ventral hernia, and it provided an adequate approach for a good repair [6, 12]. Our recurrence rate is 13.3%. Kent and colleagues are the only ones to have previously reported on the recurrence rates of diaphragmatic hernias after repair. They experienced a 29% recurrence rate; 6 of 22 from the first attempt and 2 of 6 after a second attempt. The risk of recurrence was similar whether mesh was used or not (30% versus 27%) [1]. This underscores the importance of prevention.
Complications occurred in nearly two thirds of our patients (60%). The complications led to an increased length of hospitalization when compared with those who had no complications. This compares with the previous report of a 27% complication rate by Kent and colleagues and 28.6% by Vallböhmer and colleagues. Our mortality of 0% also compares favorably with the 4.5% mortality rate of Kent and colleagues and 14.3% of Vallböhmer and colleagues [1, 2].
As previously mentioned, prevention is always the best solution. The main cause of this postoperative complication may be from the extensive blunt dissection of the hiatus during esophagectomy [12, 19]. At times it is even necessary to incise the crura, making the hiatus wide enough to allow passage of the gastric conduit. Some reports in the literature show that the only statistically significant associated risk with increased herniation was iatrogenic enlargement of the hiatus [10, 19]. Although there are reports of left, right, and anterior crural division, the literature almost uniformly supports dividing the crura anteriorly, minimizing excessive widening of the hiatus, as was done in the majority of our 15 cases [2, 6, 10, 12, 19]. One exception is Fumagalli and colleagues; they recommend a right lateral phrenotomy to avoid diaphragmatic compression of the gastric conduit after the transposed stomach has elongated [14]. It is recommended that the diaphragmatic hiatus be opened only enough to allow the conduit to pass up into the chest. If it is made too large, sutures can be placed to narrow the aperture. There are also those who are proponents of placing tacking sutures between the gastric conduit and the crura, but some argue this poses unnecessary risk to the primary vascular supply of the gastric conduit and advise against this technique [1–4, 6, 8, 12, 14, 19].
Interestingly, there are very few reports of herniation into the right chest [1, 6, 11, 14, 16]. The reason for this left predominance is unknown. It has been hypothesized that the mesentery along the lesser curvature, and the adhesions induced by the gastric staple line, prevent tissue from herniating into the right chest [1, 2]. Others have theorized that the greater curvature of the stomach is smooth and therefore has fewer adhesions than the cut surface of the lesser curvature, allowing abdominal contents to slide into the left chest [1, 2]. Another more plausible reason for the left-sided predominance is that the left and caudate lobes of the liver block access of the abdominal contents to the right chest [2]. Whatever the reason, it is striking so few right-sided hernias have been reported (Fig 3).
In summary, diaphragmatic hernias after an esophagectomy are unusual, but seem to be more common after a minimally invasive approach. The diagnosis is fairly obvious from a chest radiograph and is clearly seen with computed tomography of the chest. Most patients are symptomatic at presentation. Although we observed a substantial rate of morbidity, there were no perioperative deaths. We recommend that these hernias be repaired unless the patient's comorbidities and performance status preclude surgical intervention. We prefer to reapproximate the diaphragmatic crura without the use of mesh. Long-term outcome is usually excellent.
|
References
|
|---|
- Kent MS, Luketich JD, Tsai W, et al. Revisional surgery after esophagectomy: an analysis of 43 patients Ann Thorac Surg 2008;86:975-983.[Abstract/Free Full Text]
- Vallböhmer D, Hölscher AH, Herbold T, Gutschow C, Schröder W. Diaphragmatic hernia after conventional or laparoscopic assisted transthoracic esophagectomy Ann Thorac Surg 2007;84:1847-1852.[Abstract/Free Full Text]
- Kaushik R, Sharma R, Attri AK. Herniation of colon following transhiatal esophagectomy Indian J Gastroenterol 2005;24:122-123.[Medline]
- Terz JJ, Beatty JD, Kokal WA, Wagman LD. Transhiatal esophagectomy Am J Surg 1987;154:42-48.[Medline]
- Streitz Jr JM, Ellis Jr FH. Iatrogenic paraesophageal hiatus hernia Ann Thorac Surg 1990;50:446-449.[Abstract/Free Full Text]
- Reich H, Lo AY, Harvey JC. Diaphragmatic hernia following transhiatal esophagectomy Scand J Thorac Cardiovasc Surg 1996;30:101-103.[Medline]
- Balázs A, Forgács A, Flautner L, Kupcsulik P. A case of unusual complication of diaphragmatic herniation of the transverse colon following transhiatal esophagectomy [article in Hungarian] Orv Hetil 1997;138:2535-2538.[Medline]
- Cordero Jr JA, Moores DW. Thoracic herniation of the transverse colon after transhiatal esophagectomy J Thorac Cardiovasc Surg 2000;120:416.[Free Full Text]
- Delgado Tapia J, Ramírez Sánchez A, Molina Moreno M, Palop Manjón-Cabeza E, Ferrón Orihuela JA. Diaphragmatic hernia after transhiatal esophagectomy[article in Spanish] Rev Esp Anestesiol Reanim 2000;47:317-319.[Medline]
- Choi YU, North Jr JH. Diaphragmatic hernia after Ivor-Lewis esophagectomy manifested as lower gastrointestinal bleeding Am Surg 2001;67:30-32.[Medline]
- Granke K, Hoshal Jr VL, Vanden Belt RJ. Extrapericardial tamponade with herniated omentum after transhiatal esophagectomy J Surg Oncol 1990;44:273-275.[Medline]
- Hamaloglu E, Topaloglu S, Törer N. Diaphragmatic herniation after transhiatal esophagectomy Dis Esophagus 2002;15:186-188.[Medline]
- Aly A, Watson DI. Diaphragmatic hernia after minimally invasive esophagectomy Dis Esophagus 2004;17:183-186.[Medline]
- Fumagalli U, Rosati R, Caputo M, et al. Diaphragmatic acute massive herniation after laparoscopic gastroplasty for esophagectomy Dis Esophagus 2006;19:40-43.[Medline]
- Barbier PA, Luder PJ, Schüpfer G, Becker CD, Wagner HE. Quality of life and patterns of recurrence following transhiatal esophagectomy for cancer: results of a prospective follow-up of 50 patients World J Surg 1988;12:270-276.[Medline]
- Heitmiller RF, Gillinov AM, Jones B. Transhiatal herniation of colon after esophagectomy and gastric pull-up Ann Thorac Surg 1997;63:554-556.[Abstract/Free Full Text]
- Franceschi A, Mariette C, Balon JM, Fabre S, Triboulet JP. Diaphragmatic hernia after esophagectomy: 2 case reports and review of the literature [article in French] Ann Chir 2002;127:62-64.[Medline]
- Audebert A, Wind P, Sauvanet A, et al. Diaphragmatic hernia is a rare complication of oesophagectomy for cancer[article in French] Ann Chir 2005;130:21-25.[Medline]
- van Sandick JW, Knegjens JL, van Lanschot JJ, Obertop H. Diaphragmatic herniation following oesophagectomy Br J Surg 1999;86:109-112.[Medline]
- Katariya K, Harvey JC, Pina E, Beattie EJ. Complications of transhiatal esophagectomy J Surg Oncol 1994;57:157-163.[Medline]
- Gollub MJ, Bains MS. Herniation of the transverse colon after esophagectomy: is retrocardiac air a normal postoperative finding? AJR Am J Roentgenol 1997;169:481-483.[Abstract/Free Full Text]
- Johnson CD, Shandall A. Incisional hernia of the diaphragm causing large bowel obstruction J R Coll Surg Edinb 1987;32:51-52.[Medline]
- Ellis Jr FH, Gibb SP. Esophagogastrectomy for carcinoma: current hospital mortality and morbidity rates Ann Surg 1979;190:699-705.[Medline]
- Molina JE, Lawton BR, Myers WO, Humphrey EW. Esophagogastrectomy for adenocarcinoma of the cardia. Ten years' experience and current approach. Ann Surg 1982;195:146-151.[Medline]
- Agha FP, Orringer MB. Colonic interposition: radiographic evaluation AJR Am J Roentgenol 1984;142:703-708.[Abstract/Free Full Text]
- Agha FP, Orringer MB, Amendola MA. Gastric interposition following transhiatal esophagectomy: radiographic evaluation Gastrointest Radiol 1985;10:17-24.[Medline]
Top
Abstract
Introduction
