46: Techniques for Repair of Paraesophageal Hiatal Hernia

A multiplicity of terms has been used over the years to describe paraesophageal hernia. These reflect the considerable confusion that persists to this day concerning its pathophysiology and treatment. Terms like up-side-down stomach, rolling hernia, intrathoracic stomach, parahiatal, and paraesophageal hernia have all been used to describe this clinical condition. Any herniation of the fundus and/or body of the stomach into the chest anterior or lateral to the esophagus is considered to be a paraesophageal hernia. Once considered an immediate indication for surgical repair, the role of surgery in this operation is changing and patients receive treatment appropriate to their complaints. Persistent symptoms, in particular, those related to the mechanical effects of paraesophageal hernia, eventually will result in an indication for surgery.

The common classification for paraesophageal hernia distinguishes four types (Table 46-1). The sliding type hiatal hernia (Type I) is the most common. It occurs when the gastroesophageal junction (GEJ), along with a leading portion of the gastric cardia, slides through the esophageal hiatus into the mediastinum. The sliding hernia may progress to a rolling hernia. As the muscles of the hiatus weaken, the hiatus becomes wider, causing an ever increasing portion of the stomach along with the GEJ to slip through the opening, until the entire greater curvature rolls into the posterior mediastinum (hence the name rolling hernia). A rolling hernia is therefore seen as a progression from a small sliding hernia to complete herniation of the greater curvature of the stomach. It is now generally accepted that most paraesophageal hernias begin as sliding hiatal hernias.

In rare cases, the GEJ may remain in its normal intra-abdominal position. This is called the “true” paraesophageal hernia (Type II). Here a portion of the stomach and sometimes other parts of the abdominal viscera migrate into the mediastinum either through the esophageal hiatus or through an adjacent defect in the diaphragm.

Occasionally, the stomach may further migrate into the chest causing organoaxial volvulus, in which usually more than half the stomach lies in the mediastinum, a condition known as “giant” paraesophageal hernia (Type III). The lesser curvature remains in the abdomen, fixed by the gastrohepatic omentum and the left gastric vessels, which explains the asymmetric displacement of the stomach generally observed. With organoaxial volvulus, the fundus folds over anteriorly to the esophagus and toward the right side of the mediastinum.

Less frequently, if the defect becomes very large, colon together with omentum, spleen, and eventually small bowel may migrate into the chest, together with the stomach (Type IV).

Types III and IV are most commonly seen in elderly patients. Aging is associated with diminished muscle tone and loss of strength and elasticity of the connective tissues in the phrenoesophageal membrane, which predispose to thinning and widening of the esophageal hiatus. In addition, elderly patients often develop kyphosis. The resultant change in posture may further stretch and weaken the diaphragmatic structures. Finally, obesity can be a predisposing factor because it causes an increase of intra-abdominal pressure on the stomach.

If the herniated stomach becomes trapped between the margin of the esophageal hiatus and the esophagus, this may result in an unreducible incarcerated hernia. This condition may eventually lead to life-threatening complications, such as dilatation of the intrathoracic stomach with partial or complete obstruction, ischemia of a strangulated intrathoracic stomach with subsequent necrosis and perforation, and soiling of the mediastinum or pleural space with gastric contents resulting in infection. Localized pressure ulceration at the margin of the hiatus may result in bleeding (often occult) and anemia.

The anatomical derangements associated with the full spectrum of paraesophageal hernias are summarized in Fig. 46-1.

Medical Management

Paraesophageal hernias are usually discovered on diagnostic radiographic images in patients with symptoms of heartburn and chest pain. They also may be discovered incidentally in asymptomatic individuals. After careful questioning to rule out potential life-threatening pathology, treatment appropriate to the patient's symptoms is recommended. Patients without symptoms do not need any treatment. Patients with symptoms of reflux are treated lege artis. Moderate dysphagia can be remediated by small meals of semisolid consistency. A prokinetic, for example, metoclopramide, is helpful for treating symptoms of early satiety and fullness. The presence of moderate anemia requires iron replacement. If there is a high index of suspicion for incarceration, the introduction of a nasogastric tube will provide relief. However, persistence of symptoms, in particular, those related to the mechanical effects of paraesophageal hernia, eventually will result in an indication for surgery.

Surgical Management

In 1967, Skinner and Belsey published a classic paper¹ that drew attention to the potentially life-threatening complications of symptomatic paraesophageal hernias. In their series, over 25% of patients with a paraesophageal hernia treated with medical therapy died of catastrophic complications, chiefly, aspiration-related pneumonia. This led to the long-standing practice of treating hernias surgically, even those with minimal symptoms. That practice has been challenged over the past several decades with the advent of improved diagnostic procedures, minimally invasive approaches, and long-term outcome data from clinical studies. Even so, giant paraesophageal hernias are more complex than the garden variety sliding hernia, and this influences both surgical approach and outcome. Besides the more complex technical aspects of the surgery, itself, these interventions carry a higher rate of postoperative mortality and morbidity, even in expert hands, causing some surgeons to challenge the necessity of surgical intervention for many large or giant paraesophageal hernias.

A study by Allen et al.² from the Mayo Clinic in 1993 reported on 147 patients, 23 of which were not operated but instead followed for a median of 6.5 years. Not a single patient of this series developed life-threatening complications. In four patients, symptoms progressed resulting in elective surgery in two.

In 2002, using a Markov Monte Carlo decision analytic model, Stylopoulos et al.³ calculated that for the many patients who have only minor symptoms, such as heartburn, bloating, and so on, a policy of watchful waiting entails a lifetime risk of only 1.1% per year for the development of acute symptoms requiring surgery, with mortality related to emergency surgery of only 5.4%.³ This means that the overall lifetime risk of death is approximately 1% which is less than the reported 30-day mortality in series from high volume centers. Moreover, the older the patient, the higher the postoperative mortality. In the series published by Luketich et al.,⁴ the 30-day mortality in patients older than 80 years was 8%.

Given the fact that progression toward life-threatening symptoms/complications is low and occurs mostly in elderly populations, elderly patients with mild symptoms should not undergo surgery. Therefore, for every patient, the advantages of repair should always be carefully weighed against the risks of surgery.

An attempt should always be made to decompress the herniated stomach in patients presenting with symptoms of incarceration. This avoids the necessity of performing an emergency procedure, which is associated with high mortality, and gives adequate time, subsequently, for a thorough preoperative investigation and preparation for elective surgery under optimal conditions, for example, correction of anemia, treatment of aspiration-related pneumonia.

In cases involving clear incarceration that cannot be decompressed, gastric ischemia due to strangulation, massive bleeding, and perforation, emergency surgery is unavoidable. Depending on the experience of the surgeon, an open transthoracic or transabdominal or laparoscopic approach can be used.

The rate for conversion to open surgery, laparotomy, or thoracotomy is up to 15% in laparoscopic cases.⁵ Emergency repair in octogenarians is high (up to 16%) in the series reported by Poulose et al.⁶

Signs and Symptoms

Symptoms that accompany paraesophageal hernias are either related to reflux or to the mechanical aspects of the herniation. It is important to acquire a complete history and physical examination to elicit the typical symptoms of gastroesophageal reflux disease (heartburn and regurgitation) and dysphagia. A detailed discussion of the symptoms related to reflux is provided in Chapter 36.

Many patients with paraesophageal hernias remain asymptomatic for years or have only vague, intermittent symptoms. Although patients may initially deny symptoms, careful questioning can reveal complaints that are related to the esophagus, for example, dysphagia, epigastric or substernal pain, postprandial substernal fullness, nausea, and hiccupping.

Symptoms related to intermittent incarceration are postprandial distress or early satiety and fullness, epigastric pain, intermittent vomiting, or regurgitation. A very large volvulus may cause dyspnea, and on occasion, in this respect, atelectasis may be seen on chest CT.

Anemia and iron deficiency are usually the result of chronic occult bleeding from a pressure ulcer at the margin of the hiatus. In such cases, the patient may present with complaints of exercise-related dyspnea, fatigue, weakness, palpations, and in the case of concomitant coronary disease, angina pectoris. Painful ulceration of the herniated fundus with perforation and massive bleeding may occasionally occur. The perforation may extend into the pleura, mediastinum, or pericardium. When volvulus is present, complete obstruction of food passage may occur.

A posteroanterior and lateral chest x-ray often will show air or an air–fluid level in the mediastinum posterior to the heart indicating the presence of a herniated stomach.

Diagnostic Evaluation

The barium swallow is the gold standard diagnostic study for paraesophageal hernias (Fig. 46-2). It reveals the anatomy and relationship of the esophagus and stomach. It also will reveal whether there is free passage of contrast material (this may include a more solid contrast bolus) into the stomach. Less often, it may be helpful in locating the GEJ in relation to the diaphragm. This study reveals the anatomy and relationship of the esophagus and stomach. It also will reveal whether there is free passage of the contrast material (this may include a more solid contrast bolus) into the stomach. Often, it may be helpful in locating the GEJ in relation to the diaphragm. CT scan may be used to assess the extent of volvulus and the size of the hiatus or in identifying other abdominal organs that have migrated into the chest. Contrast studies may also reveal esophageal motility disorders, strictures, endoluminal masses, or diverticula.

Esophagoscopy is critically important; however, in cases involving a large volvulus it might be very difficult if not impossible to pass the endoscope beyond the herniated and distorted gastric pouch, into the antrum, down to the pylorus.

Endoscopy more precisely identifies the location of the GEJ in relation to the hiatus and permits evaluation for the presence and severity of esophagitis, Barrett metaplasia, or cancer to be confirmed or ruled out by biopsy. Equally important, endoscopy makes it possible to measure the distance between the GEJ and the hiatus. If this distance is >5 cm, a strong case can be made for acquired short esophagus, which calls for a different treatment approach. The patient must be awake and breathing during the endoscopy to correctly identify impingement of the hiatus.

Manometry may be useful. However, the study may be difficult to perform in patients with large hiatal hernia because of the distorted position of the stomach and GEJ. Nevertheless, through manometric studies, Maziak et al.⁷ were able to show that patients presenting with paraesophageal hernia have a significantly shorter intersphincteric length (i.e., the distance measured on manometry between the upper and lower esophageal sphincter). This has an important consequence at the time of surgical repair.

Manometry and the more recently introduced high-resolution manometry (HRM) will also provide valuable information on the motor function of the esophageal body and the function of the lower esophageal sphincter.

A 24-hour pH study and multichannel impedance pH monitoring, if technically possible, will document the presence, nature, and severity of reflux.⁸

As to the surgical approach and technique, there are no clear guidelines and several controversies persist. Many surgeons prefer laparotomy, whereas others prefer thoracotomy. Proponents of thoracotomy claim that the wide exposure afforded by this approach permits maximal mobilization of the esophagus in the event that a short esophagus is revealed at the time of surgery, or if the patient is found at surgery to have a massive nonreducible Type III or IV hernia. Since the introduction of minimally invasive surgery, the trend has been to approach these hernias laparoscopically. However, given the often complex presentation of these cases, conversion to open surgery may become necessary and the surgeon should be familiar with all surgical approaches before undertaking this repair to tailor the intervention to the patient's individual situation.

It is now well accepted that a complete excision of the hernia sac is mandatory to avoid recurrence of the hernia.^9,10 It is probably easier to excise the sac through a thoracotomy, but with increasing experience, it appears feasible to obtain complete excision through a laparoscopic approach (Fig. 46-3). Whatever the route of access, it is of paramount importance to resect the sac without damaging the vagus nerves. With the laparoscopic approach, the sac should be excised before mobilizing the esophagus and by staying away from the esophagus (Fig. 46-4A,B). This can be done by everting the sac at the beginning of the intervention and incising the sac at the border of the hiatus but avoiding damage to the crural muscle. After entering the mediastinum, the sac is further dissected and pulled down allowing both vagi to be identified. It is also important to identify the pleura to avoid inadvertent opening of the pleural sac, which may result in hemodynamic instability due to tension pneumothorax. It is also important to keep the peritoneal covering of the crura intact, or it will be difficult to obtain successful primary closure of the hiatus.

Once the hernia is reduced (Fig. 46-5), the next step is to mobilize the esophagus in the posterior mediastinum. This step is essential to achieving adequate esophageal intra-abdominal length. In particular, full mobilization up to the level of the carina is mandatory in patients who present with a suspicion of short esophagus. In some of these patients, the presumed short esophagus is simply the result of loss of elastic recoil and adequate intra-abdominal length is easily obtained. In patients with a true shortening of the esophagus, adequate length may be difficult to achieve. For these cases, it can be helpful to completely mobilize the fat pad off the stomach and the distal esophagus, again taking great care not to damage the vagus nerves (Fig. 46-6). This maneuver allows one to identify more precisely the true anatomic junction by visualizing the longitudinal esophageal fibers. It also allows one to gain some additional length, which is required for a tension-free 2-cm intra-abdominal length.

Although this intra-abdominal segment contributes to the antireflux barrier, it also provides the necessary length to perform an antireflux procedure under optimal conditions.¹¹ Indeed, given the fact that reducing a large hernia sac requires extensive esophageal dissection and causes disruption of, what is for most patients, an already deficient antireflux mechanism, significant postoperative reflux will be the most likely consequence.

If, despite all these maneuvers, the necessary 2 cm cannot be obtained, a gastroplasty is unavoidable. There are several ways to perform a gastroplasty, but today most surgeons prefer the wedge Collis gastroplasty (Fig. 46-7),¹² usually in combination with a 360-degree short floppy Nissen fundoplication (see Chapter 39). If the patient has a clear motility disorder, a partial fundoplication, for example, Toupet, is usually preferred (see Chapter 40). When an open transthoracic approach is used, and if, after full mobilization of the esophagus up to the level of the aortic arch, an insufficient length is obtained, a Collis Belsey gastroplasty is performed as described by Pearson et al.⁷ and illustrated in Chapter 38.

The use of gastropexy is controversial. Nissen¹³ and Boerema and Germs¹⁴ were the first to use gastropexy to anchor the stomach against the abdominal wall in an effort to decrease the incidence of recurrence after repair of a paraesophageal hernia. Ponsky et al. in 2003¹⁵ showed favorable results of gastropexy after laparoscopic repair of paraesophageal hernia with no recurrence at 2 years follow-up in a series of 28 patients. However, Larusson¹⁶ found gastropexy to be one of the variables associated with an increase in postoperative morbidity.

Some surgeons prefer gastropexy over antireflux procedures in frail and elderly patients because of the potential undesirable side effects of the latter that eventually may jeopardize the postoperative course. If symptomatic reflux occurs, it is hoped that this can be controlled with medical therapy (proton pump inhibitors [PPIs]).

The final but equally critical step of this intervention is the closure of the hiatus (Fig. 46-8A,B). The risk for recurrent herniation is greater with paraesophageal hernia compared with Type I sliding hernia, because of the widening of the hiatus and weakening of the muscles, especially the left crus, caused by the large herniated stomach. This has stimulated some surgeons to use prosthetic materials to minimize the risk of recurrence.¹⁷ Prosthetic material can be used as an “onlay mesh” draped over the closed pillars and around the esophagus or actually as a “curtain” incorporated in the hiatus itself. Given the presumed higher risk for erosion into the esophagus or stricture formation, most prefer to use Gore-Tex or biomaterials such as Surgisis.

Zaninotto et al.¹⁸ found a 42% recurrence rate after repair without reinforcement of the crura versus 9% with mesh reinforcement using Gore-Tex. A randomized controlled trial using the biologic prosthesis, Surgisis, showed initial superiority over closure without mesh with 9% versus 24% recurrence rates, respectively.¹⁹ This advantage appears to diminish over the long term.

For the majority of cases, if the surgeon uses meticulous technique and avoids stripping away the overlying peritoneum, the crura can be sufficiently mobilized to permit a tension-free closure.

Nason et al.¹⁰ have found it helpful to create a pneumothorax on the left side by introducing a pigtail catheter. This results in a laxity of the left pillar as the diaphragm becomes floppy. Usually three to four stitches will suffice, and to avoid too much anterior displacement, additional sutures in front of the esophagus may be necessary to obtain sufficient narrowing of the hiatus.

When the surgery is completed, most patients will be extubated without any difficulty and observed in the recovery room. However, depending on the comorbidities and/or advanced age, some patients may require transfer to the ICU for either a prolonged period of ventilation or further observation. Other aspects that may contribute to a possible admission to the ICU are nonelective technical difficulties at surgery, and/or prolonged duration of the surgery and blood loss.

In the series of 662 laparoscopic repairs published by Luketich et al.,⁴ in 29% of the cases, the surgery lasted more than 260 minutes and about one in three patients was admitted to the ICU. These figures again reflect the difficult and complex nature of paraesophageal hernia repair.

If the postoperative course is uneventful, an upper gastrointestinal (GI) contrast study with water-soluble contrast is advised to exclude the possibility of a leak on the esophagus or at the wrap, as well as the possibility of an early intrathoracic recurrence. Small contained leaks can be managed by conservative treatment, that is, NPO and antibiotics for several days until the leak has been resolved on a control contrast study. Larger leaks reflecting a larger tear as well as recurrent herniation in the chest will require reintervention. Other possible reasons for early reintervention include wound infection, paraesophageal hematoma in the residual former hernia cavity, and empyema.

In the absence of leak or herniation on the contrast study, oral alimentation is resumed on postoperative day 1 starting with liquids and soft diet after the nasogastric tube has been removed. For some patients, however, it may take longer for the stomach to regain its peristaltic activity, resulting in prolonged drainage of large amounts of gastric contents via the nasogastric tube. Such patients need to be kept NPO or only sips of water and their nasogastric tube needs to be kept in place for several days until sufficient peristaltic activity has resumed and the gastric drainage level falls below 200 cc. When the postoperative course is uneventful, patients usually can be discharged on day 2 or 3.

The most commonly encountered complication after hernia repair is dysphagia for solid foods. In most patients, this is a temporary phenomenon related to postoperative edema of the tissues resulting in a too tight wrap, but usually this will subside after 3 to 4 months. In the event of persistent dysphagia, Savary bougie, or eventually pneumatic dilatations, may become necessary to resolve the problem. In a few cases (1%–2%), redo surgery is needed to remediate the problem.⁴

A number of patients will suffer from reflux and in most of them lifelong medical therapy with PPIs will suffice to control the symptoms. In a few patients, redo surgery will be necessary but only after objective documentation of the reflux and its degree of severity.

Further long-term follow-up with quality-of-life questionnaires is advocated by most authors, such as the SF-56 (a commonly used general measure of health-related quality of life) and the Gastrointestinal Quality of Life Index (GIQLI). Over time, a risk of recurrent symptoms as well as recurrent hernia has been documented. Anatomical recurrence has been noted in 10% to 30% according to the literature, but fortunately, most of these are small and asymptomatic.

Other more rare procedure-related complications include gas bloating, postvagotomy diarrhea or dumping, and small bowel herniation in the chest causing obstruction.

Numerous articles have been published about the surgical repair of paraesophageal hernia.^20-31 The durability of the surgical repair is difficult to assess, as a consequence of inconsistent reporting of follow-up intervals and/or lack of objective data. In most published series, the follow-up interval is rather short. Some papers describe symptom control only, whereas others combine symptomatic outcome with objective assessments such as endoscopy, barium swallow, 24-hour pH monitoring, and manometry. Some studies focus solely on the incidence of anatomic recurrence. In general, open surgery, which is usually performed via transthoracic approach, produces good-to-excellent long-term results in 85% to 99% of patients, with a median follow-up of 2 to 4 years. Laparoscopic repair is associated with a higher rate of anatomic recurrence (15%–40%), but for the most part, these recurrences remain asymptomatic. Patients repaired laparoscopically rarely experience reincarceration, and if the patient does experience gastroesophageal reflux postoperatively, it is generally well managed by PPI therapy. Consequently, the incidence of reoperation with paraesophageal hernia repair is low (3%–4%) even with the laparoscopic approach. Clearly, longer and better follow-up reports are needed to assess the true success rate of surgery for these complex hernias, and definitely these problems should be dealt with by expert surgeons.

Laparoscopic surgery has revolutionized the field of hiatal hernia management and has increased the volume of cases referred to surgeons. There is still some concern about long-term recurrence, which some series have noted with the laparoscopic approach. Whether this is related to the learning curve or to fundamental shortcomings of this approach, remains to be seen. Those who apply the same principles regardless of approach—open versus laparoscopic—such as good mobilization, hiatal reconstruction with buttressing, and tension-free wraps with appropriate use of gastroplasty, have excellent results.