Pledgeted Repair of Giant Hiatal Hernia Provides Excellent Long-Term Results

Abstract

Background:

Use of mesh in hiatal hernia repairs is a topic of debate. We present our experience in laparoscopic primary (nonmesh) repair of giant hiatal hernia.

Materials and Methods:

All laparoscopic antireflux procedures done by a single surgeon from November 1997 to October 2006 were retrospectively reviewed. Inclusion criteria were primary crural closure with pledgets and giant hiatal hernia (greater than one-third of the stomach in the chest by esophagram, greater than 5 cm in length endoscopically, or greater than one-third of the stomach in the chest operatively). We attempted to reach all patients who met inclusion criteria and administered the Reflux Symptom Index (RSI) and Quality of Life Scale for Gastroesophageal Reflux Disease (QLSGR) questionnaires.

Results:

In total, 89 patients met inclusion criteria. The male-to-female ratio was 32:57. Average age was 62.7 years. Average body mass index was 29.3 kg/m². Average length of stay was 2 days, and mean clinic follow-up was 161 days. At the most recent follow-up, 62% of patients were asymptomatic. The most common postoperative symptoms were dysphagia (16%), reflux/emesis (5%), bloating (5%), nausea (4%), epigastric pain (4%), and heartburn (3%). There were six (6.7%) recurrences on esophagogastroduodenoscopy or upper gastrointestinal examination. Five patients with recurrence were symptomatic. Of the 89 patients, 29 (33%) completed the questionnaire, with a mean follow-up of 69.7 months. Average RSI score was 12 (maximum possible score, 45). In six of nine categories, the average score was less than 1 (possible score, 0–5). Average QLSGR score was 12 (maximum possible score, 45). For satisfaction with the present condition, the average score was 4.34 (maximum score, 5), and 82.7% of respondents were satisfied or very satisfied with their present condition.

Conclusions:

Laparoscopic primary repair of giant hiatal hernia provides excellent long-term results. We found that 62% of patients were asymptomatic at the last follow-up and that 82% of respondents were satisfied or very satisfied. The recurrence rate was 6.7%.

Introduction

Whether or not to use mesh to repair giant hiatal hernia defects in order to decrease the rate of recurrence is a topic of much discussion and debate among surgeons. We define giant hiatal hernia as a paraesophageal hernia in which greater than one-third of the stomach has herniated into the chest or the hernia is larger than 5 cm endoscopically. Our institution has repaired giant hiatal hernias laparoscopically with pledgeted primary crural closure for years. This method has provided good patient satisfaction and low recurrence rates. Primary repair of giant hiatal hernias avoids the complications associated with mesh repair, but the literature includes little research on the long-term follow-up of pledgeted primary repairs. Additionally, no standard method of measuring the size of the hiatal defect during surgery has been established, although Granderath¹ made a significant contribution with his measures of hiatal hernia surface area. We propose that the number of sutures used during primary crural closure, which generally correlates with the size of the hiatal defect, can be used as an indirect gauge of hiatal defect size. Our goals were to review our experience in laparoscopic primary repair of giant hiatal hernias with particular attention to postoperative symptoms and recurrence rates and to determine if the number of sutures used during crural closure correlated with hiatal hernia recurrence.

Materials and Methods

We retrospectively reviewed all laparoscopic antireflux procedures performed at the Ochsner Medical Center (New Orleans, LA) by a single surgeon from November 1997 to October 2006 (380 cases). Inclusion criteria were primary crural closure with pledgets and giant hiatal hernia, defined as having greater than one-third of the stomach in the chest according to barium esophagram, size larger than 5 cm in length according to endoscopy, and confirmed at operation by seeing that greater than one-third of the stomach in the chest laparoscopically.

Operative repair included complete removal of the hernia sac and closure of the hiatus posterior to the esophagus with 1-×1-cm Teflon pledgets on both sides of the crura 1 cm apart using 0 NUROLON^® sutures (Ethicon Endosurgery, Cincinnati, OH) intracorporeally tied. Mediastinal dissection was performed until 3–4 cm of esophagus lay without tension in the abdominal cavity. Typically formed over 60 French bougies were 2-cm fundoplications.

We reviewed records for date of birth, gender, preoperative diagnostic tests (x-rays, esophagogastroduodenoscopy [EGD], barium esophagrams, etc.), preoperative symptoms, postoperative symptoms, date of last follow-up, symptoms on last follow-up, and recurrences that were defined as greater than 2 cm of hiatal herniation on computed tomography (CT), EGD, or barium swallow. Next, we examined records for the number of sutures used in the crural closure and then attempted to reach all patients who met inclusion criteria so we could conduct a telephone interview. After patient consent was obtained, we administered the Reflux Symptom Index (RSI) and Quality of Life Scale for Gastroesophageal Reflux Disease (QLSGR) questionnaires.^2,3

Results

In total, 89 patients met inclusion criteria (Table 1). The male-to-female ratio was 32:57. Mean age was 62.7 years, mean body mass index was 29.3 kg/m², and mean length of stay was 2 days (range, 1–16 days). Mean clinic follow-up was 161 days. The most common preoperative symptoms were reflux (44%), dysphagia (26%), regurgitation/emesis (21%), chest pain (16%), heartburn (18%), epigastric pain (16%), shortness of breath (8%), nausea (8%), and food stuck in the chest (6%).

Table 1.

Demographics, Hospital Characteristics, and Available Follow-Up Information for the Entire Cohort and Patients Who Completed Questionnaires

	Entire group (n=89)	Completed questionnaires (n=29)
Male:female	32:57	7:22
Age (years)	62.7	60.6
Body mass index (kg/m²)	29.3	30.4
Length of stay (days)	2.0	1.9
Mean follow-up (days)	161	214
Recurrence	7%	7%
Symptoms at last follow-up	38%	38%

On the most recent follow-up, 62% of patients were asymptomatic (Table 2). The most common postoperative symptoms were dysphagia (16%), reflux/emesis (5%), bloating (5%), nausea (4%), epigastric pain (4%), heartburn (3%), cough (3%), chest pain (2%), and abdominal pain (2%). We found six (7%) recurrences on EGD or upper gastrointestinal series (Table 1). Mean time to recurrence was 24.5 months, and 5 patients with recurrence had symptoms on their most recent follow-up visit.

Table 2.

Patient Symptoms on the Most Recent Clinic Visit

Symptom	Percentage
Asymptomatic	63%
Dysphagia	16%
Reflux/emesis	5%
Bloating	5%
Nausea	4%
Epigastric pain	4%

Mean time to follow-up was 161 days.

Of the 89 patients, 57 had follow-up studies (64%) (CT, barium swallow, or EGD), 40 patients had multiple studies, and 35 had studies 1 year or more after surgery (39%). All 11 barium swallows done between surgery and 3 months after were negative. Between 4 and 11 months there were two recurrences, with one positive CT and one barium swallow. The other studies done during this period were two CTs, six barium swallows, and three EGDs; all were negative. In one of the cases CT was positive and EGD was negative, but the patient's hiatal hernia surgery needed redoing; therefore, the EGD was a false negative. In the group of patients with studies a year or more after surgery, there were nine additional recurrences identified (out of 26 CTs, 58 barium swallows, and 28 EGDs). There were four conflicting studies, with the CT positive in one, the barium swallow positive in two, and the EGD positive in one. When you include CT scans, 11 out of 89 patients had at least one positive study (12.3%). Looking at it another way, out of 57 patients studied, 11 patients had recurrences or follow-up studies were positive (19.3%). Certainly, follow-up studies were not routine and typically were obtained when there were unexpected postoperative symptoms.

The mean numbers of sutures used for patients with recurrence versus those with no recurrence were 4.17 (range, 1–9) and 4.33 (range, 1–20), respectively (P=.88). The mean numbers of sutures for patients with symptoms on the most recent clinic visit versus those who had no symptoms were 4.30 (range, 1–9) and 4.46 (range, 1–20), respectively (P=.82). In both comparisons the differences were not statistically significant.

Telephone interview

We attempted to reach all of the 89 patients included. We were able to contact 29 (33%), and all consented to the questionnaire. The mean follow-up was 69.7 months (range, 31–137 months) (Table 2). Age, body mass index, length of stay, recurrence rate, and percentage of postoperatively symptomatic patients for those who answered questionnaires were representative of the total group.

The mean QLSGR score was 12 (maximum possible score, 45). The questionnaire results are shown in Table 3. The individual scores ranged from 0 (never) to 5 (all the time). In three categories, the average score was greater than 1: heartburn (1.0), diet-changing heartburn (1.1), and difficulty swallowing (1.5). In six of nine categories, the mean score was less than 1. Using scores of 3 and above to define significant symptoms, 17% of respondents had significant heartburn, 17% had diet-changing heartburn, and 14% had difficulty swallowing.

Table 3.

Quality of Life Scale for Gastroesophageal Reflux Disease

	Average (SD)	% with score ≥3
How bad is the heartburn?	1.0 (1.7)	17%
Heartburn while lying down?	0.6 (1.2)	7%
Heartburn while standing up?	0.4 (1.0)	4%
Heartburn after meals?	0.8 (1.5)	10%
Does heartburn change your diet?	1.1 (1.9)	17%
Does heartburn wake you from sleep?	0.7 (1.4)	4%
Do you have difficulty swallowing?	1.5 (1.8)	14%
Do you have pain with swallowing?	0.5 (1.4)	7%
If you take medication, does this affect your daily life?	0.6 (1.5)	7%

Mean follow-up period for questionnaires was 69.7 months. The scoring scale was as follows: 0=no symptoms; 1=symptoms noticeable, but not bothersome; 2=symptoms noticeable and bothersome, but not every day; 3=symptoms bothersome every day; 4=symptoms affect daily activities; and 5=symptoms are incapacitating—unable to do daily activities.

SD, standard deviation.

The mean RSI score was 12 (maximum possible score, 45). The questionnaire results are shown in Table 4. The highest scores were excessive mucus/postnasal drip (1.7), coughing after eating or lying down (1.1), breathing difficulty (1.1), annoying cough (1.1), and difficulty swallowing solids (1.2). Again using scores of 3 and above to define significant symptoms, 24% of respondents had significant excessive mucus/postnasal drip, 14% had coughing after eating or lying down, 10% had breathing difficulty, 10% had annoying cough, and 14% had difficulty swallowing solids.

Table 4.

Reflux Symptom Index

	Average (SD)	% with score ≥3
Hoarseness	0.9 (1.6)	10%
Difficulty clearing throat	0.9 (1.7)	10%
Excess mucus or postnasal drip	1.7 (2.0)	24%
Coughing after eating or lying down	1.1 (1.8)	14%
Breathing difficulty or choking episodes	1.1 (1.7)	10%
Annoying cough	1.1 (1.6)	10%
Difficulty swallowing solids (meat, bread, pills)	1.2 (1.7)	14%
Difficulty swallowing liquids	0.6 (1.2)	4%
Sensation of lump in throat	0.8 (1.4)	7%

Mean follow-up period for respondents was 69.7 months. The scoring scale ranged from 0=never to 5=all the time.

SD, standard deviation.

In response to satisfaction with present condition, the average score was 4.34 (maximum score, 5). Of survey respondents, 83% were satisfied or very satisfied with their present condition.

Discussion

In our study of primary crural closure of giant hiatal hernias, the recurrence rate was 7%, based on EGD and barium swallow, and 23%, adding in CT results. Of the 89 patients, 62% were asymptomatic on the last clinic visit, and 83% of patients who completed the RSI and QLSGD questionnaires were satisfied or very satisfied. The number of sutures used for closure did not correlate with recurrence or symptoms on the most recent clinic visit.

When repairing a giant hiatal hernia, primary crural closure versus mesh closure remains controversial. Certainly, size as defined by number of sutures used to close a defect is somewhat imprecise, and we did not make note of the tension used to close the crura. No official standard of care has been established, although the most common practice is to use mesh selectively. According to a 2010 survey of members of the Society of American Gastrointestinal and Endoscopic Surgeons, only 8% of surgeons routinely used mesh.⁴ For those who used mesh selectively, 46% cited size of the defect as indication for mesh. In the largest series to date, Luketich et al.⁵ reviewed 662 patients who underwent laparoscopic hiatal hernia repair. Surgeons used mesh only when the peritoneum overlying the crura was compromised or when they could not close the hiatal opening without undue tension. They used mesh in 13% of cases, and recurrence was 16%.

Those who advocate mesh cite lower recurrence rates. A meta-analysis by Johnson et al.⁶ in 2006 showed that laparoscopic repair of large (>5 cm) paraesophageal hernias resulted in recurrence rates of 3% for mesh repairs and 15% for nonmesh repairs. In 2009, Soricelli et al.⁷ published a retrospective comparison of patients undergoing laparoscopic hiatal hernia repair with (1) suture, (2) mesh, and (3) suture plus superimposed mesh; they found recurrence rates of 10%, 2%, and 1%, respectively.

Despite a possibly lower recurrence rate, the benefit of routine mesh repair has been questioned. Anatomic recurrence does not correlate with symptom recurrence. In the largest series to date of laparoscopic giant paraesophageal hernia repairs, Luketich et al.⁵ found no differences in rates of symptoms between patients with or without recurrence. In a 2010 study, Parker et al.⁸ noted that most prior research shows a very low rate of symptoms in patients with nonmesh repair, even if radiologic recurrence exists. The data suggest the quality of fundoplication plays a greater role in postoperative symptoms than the presence of recurrence of herniation. Quality of fundoplication meaning that it is in the right place, not too tight or loose and of appropriate length. Moreover, Parker et al.⁸ found that for patients undergoing revisional surgery, those with mesh had an increased risk of requiring esophagogastric resection, increased dysphagia, chest pain, bloating, and longer operating times. Severe complications of mesh repair of paraesophageal hernias are well documented. Griffith et al.⁹ presented a mesh complication rate of 20%. Zügel et al.¹⁰ reported a case of mesh erosion into the abdominal aorta. Stadluhuber et al.¹¹ presented a 28-case series of mesh complications, including intraluminal mesh erosion, esophageal stenosis, and dense fibrosis. Nine of these patients required major resection, including esophagectomy, partial gastrectomy, and total gastrectomy. Only 5 patients did not require surgery.

Furthermore, recent data from Oelschlager et al.¹² suggested that lower recurrence rates in mesh repairs are only short term. Although 6-month recurrence rates were encouraging for biologic mesh, no difference in long-term (5-year) recurrence existed between mesh and primary repair. Six-month recurrence for mesh repair was 9% compared with 24% for primary repair, but 5-year recurrence for mesh repair was 59% compared with 54% for primary repair (P=.7). Surgeons initially hoped that preventing early recurrence would result in better long-term recurrence rates, but this does not appear to be the case.

We find it interesting that many surgeons use the size of the hiatal defect in deciding whether to use mesh, yet there is no standard method for measuring defect size. In 2007, Granderath¹ described the calculation of the hiatal surface area that takes into account the crural length and the distance between the right and left crura. To date, this is the most objective measure; prior to Granderath,¹ the determination of hiatal defect size was largely subjective. We propose that the number of sutures used to reapproximate the crura generally correlates with hiatal defect size and can be used as an indirect gauge of size. Intuitively this seems true, although a dedicated study comparing numbers of sutures to hiatal surface area will determine if this is the case. In our series, the number of sutures used did not correlate with recurrence or symptoms.

Our study has several limitations. First, it is a retrospective study of a single surgeon at a single institution. Second, a low percentage of patients responded to the questionnaires (likely due to Hurricane Katrina's effect on our community). Third, the validity of the phone call results is in question. Finally, the accuracy of our long-term recurrence rate would improve if all patients received postoperative radiologic studies. However, it is possible that those that had postoperative x-ray studies, particularly EGD and barium swallow, were symptomatic and that our rate of recurrence based on these studies is an overestimate. Many of those with CT results had them for what turned out to be cancers.

Conclusions

Laparoscopic primary pledgeted repair of giant hiatal hernia improves symptoms and patient satisfaction with a low recurrence rate. In our series, 63% of patients were asymptomatic on the most recent follow-up, and 82% of respondents to questionnaires were satisfied or very satisfied. The recurrence rate in our cohort was 7% by EGD or barium swallow (12.9% if CT is added). This is acceptable, compared with an overall 15% recurrence cited in the most recent literature for laparoscopic hiatal hernia repairs with selective use of mesh. The number of sutures used during crural closure did not correlate with recurrence of hernia or symptoms on most recent follow-up. According to recent literature, symptoms following repair may be determined by the quality of fundoplication rather than recurrence of herniation. Therefore, when using mesh to repair giant hiatal hernias, the potential benefit of lowering the recurrence of herniation is not necessarily outweighed by the risk of mesh complications. Other advantages of primary repair include faster operative time and lower cost.

Footnotes

Disclosure Statement

No competing financial interests exist.

References

Granderath

. Measurement of esophageal hiatus by calculation of the hiatal surface area (HAS). Why, when, how?. Surg Endosc, 2007; 21:2224–2225.

Belafsky

, Postma

, Amin

, et al. Symptoms and findings of laryngopharyngeal reflux. Ear Nose Throat J, 2002; 81:10–13.

Velanovich

, Vallance

, Gusz

, et al. Quality of life scale for gastroesophageal reflux disease. J Am Coll Surg, 1996; 183:217–224.

Frantzides

, Carlson

, Loizides

, et al. Hiatal hernia repair with mesh: A survey of SAGES members. Surg Endosc, 2010; 24:1017–1024.

Luketich

, Nason

, Christie

, et al. Outcomes after a decade of laparoscopic giant paraesophageal hernia repair. J Thorac Cardiovasc Surg, 2010; 139:395–404.

Johnson

, Carbonell

, Carmody

, et al. Laparoscopic mesh hiataoplasty for paraesophageal hernias and fundoplications. Surg Endosc, 2006; 20:362–366.

Soricelli

, Basso

, Genco

, et al. Long-term results of hiatal hernia mesh repair and antireflux laparoscopic surgery. Surg Endosc, 2009; 23:2499–2504.

Parker

, Bowers

, Bray

, et al. Hiatal mesh is associated with major resection at revisional operation. Surg Endosc, 2010; 24:3095–3101.

Griffith

, Valenti

, Qurashi

, et al. Rejection of Gore-Tex mesh used in prosthetic cruroplasty. Int J Surg, 2008; 6:106–109.

10.

Zügel

, Lang

, Kox

, et al. Severe complication of laparoscopic mesh hiatoplasty for paraesophageal hernia. Surg Endosc, 2009; 23:2563–2567.

11.

Stadluhuber

, El Sherif

, Mittal

, et al. Mesh complications after prosthetic reinforcement of hiatal closure: A 28-case series. Surg Endosc, 2009; 23:1219–1226.

12.

Oelschlager

, Pellegrini

, Hunter

, et al. Biologic prosthesis to prevent recurrence after laparoscopic paraesophageal hernia repair: Long-term follow-up from a multicenter, prospective, randomized trial. J Am Coll Surg, 2011; 213:461–468.