Paraesophageal hernia (PEH) represents 5 to 10% of all hiatal hernias. However, PEH is often associated with significant morbidity (volvulus, gangrene, perforation, bleeding) and mortality [8, 9]. Most authorities [10, 13] recommend that once the diagnosis of PEH is established, surgical repair is indicated. Elective open repair of PEH is effective but is associated with considerable postoperative morbidity [11]. Many reports in the surgical literature have demonstrated that laparoscopic repair of paraesophageal hernias (LRPEH) is feasible and effective [1, 19, 23]. Nevertheless, a common criticism has been limited patient follow-up and lack of objective studies (barium esophagram) to rule out recurrence. In addition, a recent study has raised concerns about an unacceptable radiographic recurrence rate following the laparoscopic approach [5]. We report our 6-year experience with LRPEH.

Methods

We retrospectively reviewed a computerized database of 166 patients with the diagnosis of large PEH, at two institutions, between May 1996 and August 2002. The research protocol was approved by the Institutional Review Board of each hospital. Large PEH was defined as the radiographic presence of one-third or more of the stomach in the chest cavity. Diagnosis of PEH was made with a barium esophagram in 153/166 patients (92.1%), a CT scan in 7/166 patients (4.2%), and a chest roentgenogram in 6/166 patients (3.7%). Upper endoscopy was performed at the time of surgery in all patients. The perioperative course of each patient was reviewed, including complications. Gastroesophageal reflux disease (GERD)-related symptoms, such as heartburn, dysphagia, regurgitation, chest pain, and gas bloat, were assessed utilizing visual analogue scores (VAS) on a scale from 0 to 10 (0 no symptoms, 10 severe symptoms). Patients were followed up with office visits and phone interviews. Of 166 patients, 149 (90%) were available for follow-up with VAS questionnaires at 6 months and 108 (65%) at 24 months. Of the 58 patients without 24-month follow-up, 41 patients (25%) had not completed 24 postoperative months, 8 patients (4.8%) had died, and 9 patients (5.4%) were lost to follow-up. We were successful in obtaining a postoperative barium esophagram in 120/166 patients (72%). Sixteen patients (9.6%) without a postoperative barium esophagram underwent a chest roentgenogram to assess the possibility of recurrence of the PEH.

Statistical analysis

Symptom scores were expressed as mean values ± standard deviation. Statistical significance was considered as a probability of a type I error of <5%. A two-tailed Student’s t-test was utilized to compare pre- and postoperative symptom scores (Graph Pad InStat version 3.01).

Laparoscopic repair of paraesophageal hernia (LRPEH): Technique

Following insufflation with carbon dioxide to a pressure of 15 mmHg, five trocars are placed as we have previously described [23]. Zero and 30° angled scopes are used as indicated for better visualization of the surgical field. In brief, the procedure begins with laparoscopic exploration and reduction of the herniated abdominal viscera from the intrathoracic hernia sac. After reduction of the herniated viscera and delineation of the scope of the hernia sac, several key technical maneuvers are performed to reduce the possibility of PEH recurrence and postoperative GERD. First, total excision of the hernia sac is performed to delineate the anatomic hiatal relationships. Excision of the sac also allows for reduction of the gastroesophageal junction and upper fundus into the abdomen. The second important intervention is division of the elongated short gastric vessels, which also allows for further reduction of the gastric fundus from the intrathoracic hernia sac. Dissection of the right and left crural margins is then performed with care taken to avoid injury to the vagal nerve trunks. Posterior crural approximation is accomplished after passing an esophageal 52–60 Fr Bougie into the stomach. Often excessive tension will be seen if the posterior crural approximation is extended too far anteriorly. In these circumstances, an anterior crural approximation is utilized to reduce the diameter of the esophageal hiatus. On occasion, a relaxing incision through the tendinous portion of the diaphragm is required to avoid excessive tension during the anterior crural approximation. We specifically avoid prosthetic mesh closure of the hiatal opening because of the risk of mesh erosion into the esophagus. Following crural approximation a total or partial fundoplication is usually performed based upon the preoperative assessment of the patient’s esophageal motility. If motility studies have not been performed prior to surgery, we tend to err toward a partial fundoplication in order to avoid postoperative dysphagia. Among elderly patients with significant comorbidities and absence of severe reflux symptoms, we have chosen to employ diaphragmatic gastropexy as opposed to fundoplication. The gastropexy is performed following reduction of the herniated viscera, hernia sac excision, short gastric vessel ligation, and crural approximation.

Results

A total of 166 patients underwent LRPEH. The mean age was 68 years (range: 30–91 years) and the male/female ratio was 0.4/1. PEH were classified based on preoperative barium esophagram studies and intraoperative findings as follows: type II: 43 (26%), type III: 104 (63%), and type IV: 19 (11%). A fundoplication was performed in 152 patients (91.6%): Nissen in 127 patients (76.5%), Toupet in 23 patients (13.8%), Dor in one patient (0.6%), and Nissen-Collis in one patient (0.6%). Fourteen patients (8.4%) underwent a gastropexy. Operative time was 160 min (range: 50–325 min). Mean hospital stay was 3.9 days (range: 1–74 days). Two patients (1.2%) were converted to open technique because of difficult hernia sac dissection and unclear anatomy.

Fourteen patients (8.4%) developed 15 perioperative complications (Table 1). One patient with perioperative esophageal leak and mediastinitis was treated with video-assisted thoracoscopic drainage with good results. Pre- and postoperative symptom scores were available in 90% of patients at 6 months and 65% of patients at 24 months. At 6 and 24 months postoperatively there was statistically significant improvement in the mean VAS for heartburn, regurgitation, dysphagia, chest pain, and gas bloat (Table 2).

Table 1 Perioperative complications
Full size table
Table 2 Visual analogue symptom scores (mean ± SDV)
Full size table

A routine postoperative barium esophagram study was obtained in 120 patients (72%), at a mean of 15 months postoperatively. Thirty-four patients (28%) had abnormal barium esophagram findings: 24 patients (20%) had a small type I sliding hernia, six patients (5%) had recurrent PEH, and four patients (3%) had wrap failure. Sixteen patients without a postoperative barium esophagram underwent a chest roentgenogram, at a mean of 7 months postoperatively. There was no evidence of PEH recurrence seen in these evaluations. Only six of the 14 patients treated with gastropexy had a follow-up barium esophagram, and therefore no valid comparison could be made with the majority of patients who underwent a fundoplication. However, none of the six patients with postoperative barium esophagram had evidence of PEH recurrence.

Four patients with small asymptomatic recurrent PEH have not undergone operative repair and remain asymptomatic at a mean follow-up of 14 months. Only two of the 24 patients with postoperative small sliding hernias required reoperation. These two reoperations were necessary to manage significant postoperative symptoms. We further evaluated the 24 patients identified with small sliding hernias on barium esophagrams following repair of their PEH. Eighteen of the 24 patients (75%) had available pre- and postoperative symptom scores for comparison. There was significant improvement in their symptomatology compared to the preoperative values (Table 3). Furthermore, these patients had a similar postoperative symptom improvement compared to the group of patients without a sliding hernia (Table 4).

Table 3 Pre- and postoperative symptom scores in patients with postoperative sliding hernias
Full size table
Table 4 Postoperative symptom scores of patients with and without sliding hernias
Full size table

A total of 10 patients (6%) required a reoperation at a median time of 7 months (range 2–26 months) following the primary operation. Two patients had large symptomatic recurrent PEH and underwent successful laparoscopic repair. In both patients the fundic wrap had migrated to the mediastinum through a partially disrupted hiatal closure. Four patients with significant GERD symptoms underwent laparoscopic reoperation; a disrupted wrap was found in three patients and a small hiatal hernia in one patient. Four patients with persistent dysphagia despite endoscopic dilations required operative treatment with a thoracoscopic approach (n = 1), open laparotomy (n = 1), and attempted laparoscopic approach (n = 2). Both attempted laparoscopic reoperations for dysphagia led to severe gastroesophageal injuries requiring a transhiatal esophagectomy in one patient and a jejunal interposition in the other patient. In both cases, there was significant scarring around the gastroesophageal junction most likely due to the repeated endoscopic dilations. In none of the reoperative cases was there evidence of a short esophagus.

Discussion

Most surgeons recommend repair of PEH, regardless of symptomatology, for patients who are good candidates for surgery [4, 15, 24]. This prophylactic approach is based on classic studies published by Skinner, Belsey, and Hill that reported a high incidence of life-threatening complications associated with PEH [6, 20]. Recently, the role of nonoperative management in asymptomatic patients with PEH was revisited [21]. In this series, the authors reported a successful “watchful waiting” approach in a hypothetical cohort of patients with minimal or no symptoms. In our series, the vast majority of patients reported at least moderate symptoms related to their PEH (i.e., chest pain, dysphagia, and significant regurgitation).

Surgical repair of PEH is highly effective but not risk-free. Complications of surgery include visceral injury, vagal nerve injury, pneumothorax, hemorrhage, pulmonary complications, and recurrent gastric volvulus [18, 22]. Laparoscopic-assisted endoscopic reduction of the herniated stomach and gastropexy with a percutaneous endoscopic gastrostomy (PEG) may be an option for patients who are unfit to undergo surgery [7].

LRPEH is an attractive minimally invasive surgical approach that offers superior visualization of the hiatal anatomy, which is crucial for the mediastinal mobilization of the esophagus. Moreover, laparoscopy is associated with a decreased physiologic insult [14]. Laparoscopic repair is also known to be effective in controlling symptoms associated with PEH [1, 10]. Despite the aforementioned advantages of the laparoscopic technique, there are concerns regarding the management of large paraesophageal hernias. Laparoscopic repair of large PEH is technically challenging because of anatomic distortion and requires meticulous dissection of the hernia sac [12, 17, 23]. Six percent of our patients required a second procedure to treat recurrent hiatal herniation or recurrent GERD symptoms. The rate of reoperation following laparoscopic PEH repair varies from 1.5% to 9% [2, 3, 4, 12, 16, 19, 22]. In our experience, reoperative surgery for dysphagia in patients with multiple endoscopic dilations is associated with a high risk of gastroesophageal injury and should be approached with the open technique.

In a recent study, the radiographic recurrence rate was unacceptably high (42%) and the authors questioned whether laparoscopic repair of large PEH is appropriate [5]. Of interest, anatomic recurrence was defined as any evidence of stomach herniation above the level of the diaphragm, including the presence of type I hernias. Patients with PEH have a large esophageal hiatus and, consequently, it is not surprising to find a small sliding hernia on postoperative barium esophagram. It is difficult to clearly estimate whether this finding is significantly higher following laparoscopic repair when compared to open repair because most of the open series lack long-term follow-up. We were able to obtain objective follow-up with a barium esophagram in 72% of our patients at a mean of 15 months. Although not perfect, this rate is satisfactory considering that only a small number of published studies of laparoscopic or open PEH repair have objective long-term follow-up [5]. The elderly nature of the patients with PEH seen in our practice and the geographic distance from the primary referral creates a hurdle for the performance of routine radiographic studies in the late postoperative period.

The question remains whether the occurrence of small postoperative sliding hernias represents an important clinical entity following repair of large PEH. Most of our patients identified with small sliding hernias have remained asymptomatic in long-term follow-up. Evaluation of symptom scores among these patients also reveals an equivalent and significant improvement compared to patients without postoperative sliding hernias. The reoperation rate among these pts is low; however, the role of esophageal lengthening procedures at the time of the primary repair in reducing the likelihood of such postoperative hernias is yet to be defined.

Conclusions

Laparoscopic approach to the management of large PEH is safe and effective. The majority of patients achieve significant improvement in preoperative GERD symptoms with a low incidence of recurrent PEH. Most small PEH recurrences can be managed conservatively. For patients requiring revisional surgery, laparoscopic reintervention is technically challenging but usually successful.