Laparoscopic Nissen Fundoplication Versus Thal Fundoplication in Children: Comparison of Short-Term Outcomes

Abstract

Background:

The aim of this study was to compare short-term outcomes, including intra- and perioperative complications following laparoscopic Nissen versus Thal fundoplication.

Patients and Methods:

From July 1998 until April 2007, 175 patients were recruited. Patients were prospectively randomized to either a Nissen wrap or a Thal wrap. Observation period was 6 weeks after surgery.

Results:

89 Nissen and 86 Thal were performed. The mean age at the time of operation (OP) was 5.2 years. Demographics were similar, although weight at OP was significantly less in the Nissen group. Intraoperative complications during a Nissen included bleeding from a liver laceration in 2 patients (1 required conversion) and small bowel perforation during open port insertion in 1 patient. There were two conversions in the Thal group, due to bleeding from the omentum in 1 patient and equipment failure in the other. In a third patient the colon was perforated during insertion of percutaneous endoscopic gastrostomy (PEG) and repaired laparoscopically. Post-OP dysphagia was similarly distributed among both groups, but was significantly more severe after a Nissen (P = 0.018). There were two early deaths: in the Nissen group, 1 child died from peritonitis after the gastrostomy tube fell out, whereas one death in the Thal group was caused by respiratory failure associated with the patient's underlying condition.

Conclusions:

There was no statistical difference in the short-term outcomes between laparoscopic Nissen and Thal fundoplication, apart from a higher rate of esophagoscopy for severe dysphagia in the Nissen group. The higher number of postoperative complications in the Nissen group was largely due to gastrostomy-related problems.

Introduction

The reported benefits of laparoscopic fundoplication compared with open surgery include less postoperative pain, faster recovery, shorter hospital stay, and better cosmesis.¹ There is now a general consensus that laparoscopy is the preferred approach to fundoplication. However, there is no consensus as to what is the most effective type of fundoplication.

The Nissen fundoplication is the most widely performed fundoplication,² but it is associated with a high incidence of postoperative dysphagia^3,4 and an appreciable recurrence rate in children with disabilities.⁵ Consequently, some authors have advocated a partial wrap. The Thal fundoplication⁶ was popularized in children by Ashcraft et al.⁷ and is used by many surgeons today.

The purpose of this report (part of a long-term prospective study) was to compare short-term outcomes of laparoscopic Nissen with laparoscopic Thal fundoplication in children. Parameters assessed included intraoperative events (conversions, complications, and time taken), immediate postoperative outcomes (need for admission to the high dependency unit (HDU) or pediatric intensive care unit (PICU), opiate requirements, time to full feeds, and length of hospital stay), and later problems up to 6 weeks (dysphagia, late postoperative complications, and deaths). Long-term follow-up results have been reported separately.

Methods

Patients and demography

Between July 1998 and April 2007, 175 patients were recruited into this prospective, randomized study at the Department of Paediatric Surgery, John Radcliffe Hospital Oxford. All patients were ≤21 years old, with an average age of 5.2 years at the time of surgery. There were 26 children (14.9%), who were 1 year of age or less. The youngest infant in our study underwent a fundoplication at 5 weeks and weighed 2.4 kilograms. Age and sex distributions were similar in both groups, but weight at the time of operation was significantly lower in the Nissen group (P = 0.036) (Table 1). One hundred twenty-one patients (69.1%) had a variety of underlying neurological disorders, 13 had esophageal pathologies (e.g., congenital diaphragmatic hernia or esophageal atresia), and 41 were classified as normal children without other significant medical conditions contributing toward the gastroesophageal reflux (GOR).

Table 1.

Demography

Total (n = 175)	Nissen (n = 89)	Thal (n = 86)	Statistics
Male (n = 100)	47 (52.8%)	53 (61.6%)	NS
Female (n = 75)	42 (47.2%)	33 (38.4%)	NS
Mean age (OP) 5.2 years (±4.7)	4.7 years (±4.8)	5.7 years (±4.7)	NS
Weight (OP) 16.4 kg (±11.7)	14.2 kg (±9.4)	18.7 kg (±13.3)	P = 0.036

OP, at time of operation; NS, not significant.

Study design

Inclusion criteria for entry into the study included proven GOR unresponsive to medical treatment, patients who had failed medical treatment or who had serious complications (i.e., apnea, aspiration pneumonia, esophagitis, or failure to thrive), or those who had a hiatus hernia. GOR was confirmed by a combination of 24-hour pH monitoring, upper gastrointestinal swallow and meal, and/or esophago-gastro-duodenoscopy (OGD).

Patients were excluded if they had had previous antireflux surgery or previous open abdominal surgery, if the parents had a preexisting preference for one of the procedures, or if parents declined to participate in the study.

Ethics approval for the study was obtained from the Oxfordshire Research Ethics Committee in 1998 (No. 04.OXA.18-1998) and subsequently renewed in 2004 and 2007.

On the day of admission, informed consent was obtained from the parents and the patients were randomized to either undergo a laparoscopic Nissen fundoplication or a laparoscopic Thal fundoplication.

For the purposes of this report, short-term follow-up lasted into the early convalescent period only (6 weeks after surgery). Outcomes assessed included intraoperative problems (conversions, complications, and time taken); postoperative recovery (need for HDU/PICU, opiate requirements, time to full feeds, and length of hospital stay), and postoperative complications, including dysphagia and early death.

Surgical procedure

In each child a nasogastric tube was inserted and the patient was placed in a modified Lloyd-Davies, head-up position with the surgeon at the lower end of the table between the patient's legs. A 5- or 10-mm 30° laparoscope was inserted through a periumbilical port. Further, 3- or 5-mm ports were placed in the right and left upper quadrants, and the liver was retracted using a Nathanson retractor (Cook^® Medical). After minimal dissection of the lower oesophagus, hiatal repair was performed by applying 1–2 crural stitches. The only difference between the two techniques was the way the fundoplication was fashioned:

In the Nissen fundoplication a 2–3-cm fundal wrap was created with 3–4 2/0 Ethibond sutures (fundus to fundus) after an initial stitch from the fundus to the oesophagus.

For the Thal fundoplication⁶ the fundus was sutured anteriorly to the oesophagus in an inverted U-pattern using 7–9 (depending on patient size) 2/0 Ethibond sutures, creating a 270° anterior wrap.

When necessary, a laparoscopically guided gastrostomy button was inserted at the end of the procedure. All operations were performed or supervised by the senior author (H. W. Grant).

Statistical analysis

Variables were compared using the 2-tailed Student t-test or chi-square test where appropriate. Significance was defined as P-value ≤0.05. Data were expressed as median values (range) or mean (±SD) as stated. SPSS software version 11.5 for Windows was used for statistical analysis.

Results

A total of 175 patients were recruited to the study; 89 underwent a Nissen fundoplication, and 86 a Thal fundoplication. In 60% of children (62 in Nissen group; 43 in Thal group) an additional laparoscopic-assisted gastrostomy was performed.

Intraoperative problems occurred in a total of 6 patients (3.4%). In the Nissen group there were three complications: 2 patients developed bleeding from a liver laceration. Neither of them required transfusion, but 1 required conversion to open surgery. The third patient had a small bowel perforation during open port (Hasson) insertion. This was identified and repaired at the time through the umbilical port.

In the Thal group there were three complications: two conversions—bleeding from the omentum (previous surgery) in 1 patient and due to equipment failure in the second. In a third patient the colon was perforated during insertion of percutaneous endoscopic gastrostomy (PEG) and repaired laparoscopically with no deleterious consequences.

There was no significant difference between the fundoplication techniques with regard to duration of surgery, need for admission to a high care unit (HDU or PICU), requirement for postoperative parenteral opiates, time to full feeds, or the length of hospital stay before discharge from hospital (Table 2).

Table 2.

Comparison of Perioperative Outcomes Between Laparoscopic Nissen and Thal Fundoplication

Total (n = 175)	Nissen (n = 89)	Thal (n = 86)	Statistics
OP-Time (minutes)^a	102.0 (±31.6)	109.1 (±33.3)	NS
PICU/HDU (n = 35)	19 (21.3%)	16 (18.6%)	NS
Parenteral opiates (n = 107)	54 (60.7%)	53 (61.6%)	NS
Time to full feeds (hours)	44.5 (±18.2)	45.6 (±23.7)	NS
Length of stay (days)	3.8 (±3.2)	3.4 (±2.1)	NS

Time for gastrostomy placement not included.

Postoperative complications were noted in 20 patients—15 in the Nissen group (16.9%) and 5 in the Thal group (5.8%). The majority (n = 16) were related to problems with the gastrostomy. Five of these patients required subsequent surgery. In 3 children the gastrostomy tube leaked and caused erosions of the surrounding skin. In the other 2 the gastrostomy tube fell out (balloon failure), resulting in peritonitis.

Other postoperative problems included wound infections in 2 (1 in Nissen group and 1 in Thal group), treated with antibiotics. There was one port-site wound dehiscence (Nissen) that needed resuturing, and 1 child with gross abdominal distension (Nissen) in the early postoperative period who had a relook laparoscopy (no problem identified) and subsequently settled (Table 3).

Table 3.

Early Postoperative Complications

Complication	Nissen (n = 89)	Thal (n = 86)
Gastrostomy site infection (n = 11)	8	3
Gastrostomy break down/peritonitis (requiring surgery) (n = 5)	4	1
Wound infection (n = 2)	1	1
Others (n = 2)	2	—
Total (n = 20)	15 (16.9%)	5 (5.8%)

Early postoperative dysphagia occurred in 12 children after Nissen fundoplication (6 had neurological problems; 6 had no neurological problems) compared with 10 after Thal fundoplication (2 had neurological problems; 8 had no neurological problems). However, severe dysphagia (meriting OGD ± dilatation) was significantly more common in the Nissen group (n = 9) than in the Thal group (n = 1) (P = 0.018) (Table 4). Looking at subgroups: 5 out of 20 (25%) normal children who had a Nissen developed severe dysphagia requiring OGD ± dilatation compared with 1 out of 34 (2.9%) in the Thal group. In the neurologically impaired group 4 out of 69 (5.8%) who had a Nissen and none out of 52 patients (0%) after a Thal had significant dysphagia.

Table 4.

Comparison of Postoperative Dysphagia and Early Death After Laparoscopic Nissen and Thal Fundoplication

Total (n = 175)	Nissen (n = 89)	Thal (n = 86)	Statistics
Dysphagia (n = 22)	12 (13.5%)	10 (11.6%)	NS
Severe Dysphagia (requiring Endoscopy) (n = 10)	9 (10.1%)	1 (1.7%)	P = 0.018
Early death (n = 2)^a	1 (1.1%)	1 (1.7%)	NS

One death in the Nissen group was related to surgery.

There were two early deaths: 1 child (Nissen) developed peritonitis after the gastrostomy tube fell out 3 days postoperatively and was mistakenly replaced into the peritoneal cavity. After relaparotomy she progressed to multiorgan failure, required extracorporeal membrane oxygenation (ECMO), and died 6 weeks after surgery. Another child in the Thal group with severe neurological problems from birth asphyxia died at home as a result of respiratory failure 5 weeks after the fundoplication.

Discussion

Laparoscopic fundoplication is commonly performed for the treatment of GOR in children. However, there is no agreement among surgeons as to what is the best antireflux procedure. This is not surprising as fundoplication is a mechanical solution for what is frequently a functional problem (e.g., often secondary to underlying neuromuscular dysmotility).

To date, there has been no prospective, randomized study comparing different laparoscopic fundoplication techniques in children. Most of the studies only report short-term results, and most of the studies involved multiple surgeons and, in some cases, multiple institutions. To overcome these problems we performed a prospective, randomized, controlled study at a single center, supervised by a single surgeon performing a standardized procedure—the only variable was the design of the wrap.

The Nissen fundoplication is the most commonly performed procedure.² The reason for choosing the Thal fundoplication as a comparison was because the Thal is a partial wrap, it is widely performed, and it was reported to be a simple and safe operative treatment for GOR in children with fewer perioperative complications and better long-term outcomes.^7,8

Fundoplication is a major surgical intervention. Although laparoscopy has reduced the surgical trauma of the procedure itself, it still carries an appreciable complication rate, particularly when combined with a gastrostomy.^9,10 Arguably, laparoscopy may be associated with fewer intraoperative problems as there is better observation, magnification of important structures, and contemporaneous peer review of the procedure via the theater monitors, whereas in open surgery only the surgeons can usually see what they are doing, and visibility is often limited.

In this series 6 patients had intraoperative problems (3.4%). None of the complications was related to the dissection of the crura or the formation of the wrap, so this study did not show any difference in the difficulty between the two techniques. Three patients (1.7%) required conversion to an open procedure. Of the six important complications, three were managed adequately laparoscopically—one liver bleed settled with topical Spongostan^® (Ethicon^®). The injury to the colon during PEG trocar insertion occurred because of dazzle from the endoscope during trocar insertion—this has not happened again since the PEG trocar is placed in the stomach under laparoscopic guidance before insertion of the flexible endoscope for the PEG. The small bowel was inadvertently picked up during open Hasson entry—this has not happened again since the peritoneum has been picked up separately. Three patients required conversion—the reasons for conversion included bleeding from the liver (1 patient) caused by the liver retractor. We have since stopped using the fan retractor and only use a Nathanson. Since then there has been no conversion due to liver trauma. In another patient there was bleeding from omentum (adhesions from previous surgery). Our conversion rate (1.7%) is comparable with the literature, ranging from 0% to 8%.^1,2,9–18

Most authors report a higher number of intraoperative problems compared with the actual conversion rate.^1,15–17 This is an indication that increasing surgical expertise and confidence means that the surgeon is able to cope with intraoperative problems laparoscopically (e.g., to stop bleeding, free adhesions, and to repair a perforation) rather than to convert to open surgery. This trend was also noted in this study.

In this study the procedures only differed in the formation of the wrap—all other parts of the surgery were standardized. In a Nissen there is the need to dissect a window behind the oesophagus to adequately mobilize the fundus and, in some cases, to divide the short gastric vessels. These steps are not necessary in a Thal fundoplication; however, extra time is taken due to the greater number of sutures in the Thal wrap. The procedures were timed from skin incision to skin closure. The extra time for placing a gastrostomy was documented but not counted toward the fundoplication. Overall, there was no significant difference in the time taken for the two procedure and the average time of 105 minutes compared favorably with the literature ranging from 55 to 185 minutes.^{1,2,10,14–16,19} Like other series there was a trend toward shorter operative times at the end of study.^1,19

Not surprisingly, there was no difference in the need to observe patients postoperatively on HDU/PICU between the two groups—31 out 35 children who required HDU/PICU had underlying neurological problems with other major comorbidities and were equally distributed between Nissen and Thal patients.

Similarly, there was no difference in the length of hospital stay between the two groups. The length of stay was related to their other morbidities or postoperative complications rather than the type of fundoplication. The time in hospital was comparable with that of other studies ranging from 1.6 to 9.3 days,^{1,2,9,11,12,17,19} with a trend toward a longer stay, the more patients had associated neurological problems.^17,20

The gastrostomy was the cause of most of the complications (9.1%), including 5 children who required further surgical intervention for a wound break down and/or peritonitis. There were more complications in the Nissen group (n = 15) than in the Thal group (n = 5) because there were 62 gastrostomies performed in the Nissen group compared with 43 in the Thal group. Gastrostomy-related complications are common in the literature with a complication rate of up to 12%.¹⁰

There was one port-site hernia. It is sometimes difficult to close 3- and 5-mm port sites at the end of the operation, as the skin incision is small and sometimes the muscle defect has slid away from the skin wound. To overcome this problem, we place our closing sutures at the time of port insertion when the muscle defect can easily be seen. This has reduced the incidence of this problem.

Dysphagia is an important side effect of laparoscopic antireflux surgery.²¹ The etiology of the postoperative dysphagia is multifactorial: the type of wrap, the degree of wrap, the tightness of the wrap, torque of the wrap, apposition of the crura, and postoperative edema. In this series, 22 patients (12.6%) experienced at least one episode of dysphagia within 6 weeks after surgery (12 in Nissen group and 10 in Thal group). In the majority of cases this was only a temporary problem and the symptoms resolved with time. This phenomenon has been described in the literature.^2,17,22 Mathei et al.¹⁷ suggested that postoperative edema around the lower oesophagus subsequently settles. In this study 9 out of 12 patients in the Nissen group experienced sufficiently severe dysphagia that required OGD ± dilatation compared with only 1 out of 10 in the Thal group. Of note, severe dysphagia was more often recognized in normal children than in children with other significant neurological disorders. We suspect that neurologically handicapped children experience this side effect as commonly but cannot communicate the symptoms.

Dysphagia in children, particularly in those with underlying neurological problems, is undoubtedly under-reported in the literature: the dysphagia rate after laparoscopic fundoplication in the pediatric population ranges from 1.6% to 31%.^9,12,15–19 Chung et al.²² reported a slightly higher rate of dysphagia after a laparoscopic Toupet operation (2.3%) compared with 1.1% after a Nissen procedure, whereas Esposito et al.² observed a trend, although not statistically significant, toward a higher rate of dysphagia after a laparoscopic Nissen (4.4% after Nissen versus 2.1% after Thal and Toupet).

There are a number of possible reasons why the Nissen fundoplication resulted in more severe and persistent dysphagia in this series. A Nissen fundoplication is a complete 360° wrap, so it is more obstructive than a Thal (which is a partial 270° anterior wrap). In a Nissen fundoplication, the posterior wrap displaces the oesophagus anteriorly just above the gastro-esophageal junction—this does not occur in a Thal fundoplication. There may also be a degree of torque in a Nissen if the fundus is not adequately mobilized, but this maneuver is not performed in a Thal fundoplication. It is also possible to fashion too tight in a Nissen fundoplication—this never happens with partial wraps. Additionally, if there is postoperative edema, it could have more impact when the wrap is circumferential.

There were two early deaths: a 5-year-old boy (an ex-premature with birth asphyxia and severe neurological problems) died at home 5 weeks after surgery because of respiratory problems not related to surgery. The second child died as a direct result of a postoperative complication (displaced gastrostomy). She was a 6-month-old baby girl who had congenital heart disease and an abnormal phenotype. Her gastrostomy fell out 3 days postoperatively and it was replaced into the peritoneal cavity. She developed peritonitis and septicemia and died on ECMO 6 weeks after surgery. The tube used in this patient was a balloon gastrostomy (Dilation Gastrostomy Kit; Medicina Ltd.). The balloon deflated and caused the tube to fall out. Whenever possible we insert PEG tubes (Corflo^® PEG Kit; Merck Serono Ltd.) under laparoscopic guidance—none of these tubes has fallen out in this series. However, in small patients PEG insertion is difficult (due to the size of the gastroscope and PEG flange), so we still insert a laparoscopy-assisted balloon gastrostomy. These patients are very closely monitored postoperatively to detect any problems as early as possible. If the tube falls out, correct intragastric positioning is confirmed radiologically before use. Three other studies in the pediatric population^9,12,20 reported also early deaths that were related to gastrostomy complications. These are a cause of morbidity and extra vigilance is required.

Summary

There was no statistical difference in short-term outcomes between laparoscopic Nissen and Thal fundoplication apart from the higher rate of esophagoscopy with or without dilatation for severe dysphagia in the Nissen group. The most frequent source of postoperative complications was caused by gastrostomy-related problems.

Footnotes

Disclosure Statement

No competing financial interests exist.

This study was presented at the Annual Meeting of the British Association of Paediatric Endoscopic Surgeons 2009, Sheffield, England, United Kingdom.

References

Rothenberg

. Experience with 220 consecutive laparoscopic Nissen fundoplications in infants and children. J Pediatr Surg, 1998; 33:274–278.

Esposito

, Montupet

, van der Zee

et al. Long-term outcome of laparoscopic Nissen, Toupet, and Thal antireflux procedures for neurologically normal children with gastroesophageal reflux disease. Surg Endosc, 2006; 20:855–858.

Fernando

, Luketich

, Christie

et al. Outcomes of laparoscopic Toupet compared to laparoscopic Nissen fundoplication. Surg Endosc, 2002; 6:905–908.

Strate

, Emmermann

, Fibbe

et al. Laparoscopic fundoplication: Nissen versus Toupet two-year outcome of a prospective randomized study of 200 patients regarding preoperative esophageal mobility. Surg Endosc, 2008; 22:21–30.

Smith

, Biemann Othersen

, Gogan

et al. Nissen fundoplication in children with profound neurologic disability. Ann Surg, 1992; 214:654–659.

Thal

. A unified approach to surgical problems of the esophagogastric junction. Ann Surg, 1968; 168:542–550.

Ashcraft

, Goodwin

, Amoury

et al. Thal fundoplication: a simple and safe operative treatment for gastroesophageal reflux. J Pediatr Surg, 1978; 13:643–647.

Van der Zee

, Rövekamp

, Pull ter Gunne

et al. Surgical treatment of reflux esophagitis: Nissen versus Thal procedure. Pediatr Surg Int, 1994; 9:334–337.

Meehan

, Georgeson

. Laparoscopic fundoplication in infants and children. Surg Endosc, 1996; 10:1154–1157.

10.

Hopkins

, Stringel

. Laparoscopic Nissen fundoplication in children: a single surgeon's experience. JSLS, 1999; 3:261–266.

11.

Allal

, Captier

, Lopez

et al. Evaluation of 142 consecutive laparoscopic fundoplication in children: effects of the learning curve and technical choice. J Pediatr Surg, 2001; 36:921–926.

12.

Capito

, Leclair

, Piloquet

et al. Long-term outcome of laparoscopic Nissen-Rossetti fundoplication for neurologically impaired and normal children. Surg Endosc, 2008; 22:875–880.

13.

Diaz

, Gibbons

, Heiss

et al. Antireflux surgery outcomes in pediatric gastroesophageal reflux disease. Am J Gastroenterol, 2005; 100:1844–1852.

14.

Dick

, Potts

. Laparoscopic fundoplication in children—an audit of fifty cases. Eur J Pediatr Surg, 1999; 9:286–288.

15.

Esposito

, Montupet

, Reinberg

. Laparoscopic surgery for gastroesophageal reflux disease during the first year of life. J Pediatr Surg, 2001; 36:715–717.

16.

Liu

, Lin

, Statter

et al. Laparoscopic Nissen fundoplication without division of short gastric vessels in children. J Pediatr Surg, 2006; 41:120–125.

17.

Mathei

, Coosemans

, Nafteux

et al. Laparoscopic Nissen fundoplication in infants and children: analysis of 106 consecutive patients with special emphasis in neurologically impaired vs. neurologically normal patients. Surg Endosc, 2008; 22:1054–1059.

18.

Steyaert

, Al Mohaidly

, Lembo

et al. Long-term outcome of laparoscopic Nissen and Toupet fundoplication in normal and neurologically impaired children. Surg Endosc, 2003; 17:543–546.

19.

Van Niekerk

, Visser

, Venter

. Laparoscopic antireflux surgery in infants and children. S Afr Med J, 2005; 95:675–677.

20.

Esposito

, van der Zee

, Settimi

et al. Risks and benefits of surgical management of gastroesophageal reflux in neurologically impaired children. Surg Endosc, 2003; 17:708–710.

21.

Pimpalwar

, Najmaldin

. Results of laparoscopic antireflux procedures in neurologically impaired children. Semin Laparosc Surg, 2002; 9:190–196.

22.

Chung

, Georgeson

. Fundoplication and gastrostomy. Semin Pediatr Surg, 1998; 7:213–219.