Introduction

Fundoplication (FP) is the antireflux surgery treatment of choice for gastroesophageal reflux disease (GERD) refractory to medical therapy in children and young adults (Figs. 1, 2, and 3). It leads to effective symptom control, reduction of GERD medication, and improved postoperative quality of life for patients [1,2,3,4,5,6,7]. Previous abdominal surgery (PAS) can make FP technically more challenging for the surgeon, mostly due to intrabdominal adhesions that lead to suboptimal intraoperative line of sight [8,9,10]. The impact of PAS on these patients is debated. It was previously reported that PAS can be associated with prolonged operative times, increased conversion and complications rates [11, 12], and poorer postoperative outcome [11,12,13,14]. However, others report comparable morbidity and functional results with and without PAS [8, 15,16,17]. The aim of the present study was to analyze the impact of PAS on FP-related short- and long-term outcome in a cohort of patients < 23 years of age following FP for medically refractory GERD.

Fig. 1
figure 1

Dissection of the hiatus (a) and abdominalization of the esophagus (b). Lv left lobe of the liver, Es esophagus, Cr r right crus of diaphragm, Cr l left crus of diaphragm, Hi hiatus esophagus

Full size image
Fig. 2
figure 2

Dorsal suture of the hiatus. Cr r right crus of diaphragm, Su dorsal suture of the hiatus, Es esophagus

Full size image
Fig. 3
figure 3

Fundoplication wrap: Formation of a 360° cuff (a) and finalizing of the cuff suture (b). Ni Nissen fundoplication, Cr r right crus of diaphragm, Es esophagus

Full size image

Material and methods

Study population and data acquisition

The study protocol was approved by the local ethics committee (S-272-2019 and S-638/2011). We retrospectively analyzed patients (age < 23 years) following FP for medically refractory GERD at our university center for pediatric surgery from February 1999 to October 2019. Study inclusion criteria were follow-up longer than 6 months after FP and the availability of complete study data. Study exclusion criteria were patients with previous solid-organ transplantations.

Study data were retrieved from the electronic patients’ charts and included demographic data (gender, age at surgery, weight at surgery, height at surgery), preoperative data (diagnosis, comorbidities, previous abdominal operation), perioperative data (operative time, type of operation [laparoscopic vs. open operation], need for conversion to open surgery, type of FP [Nissen or Thal], occurrence and grade of postoperative surgical complications according to the ‘Clavien–Dindo’ classification [18]) and outcome data (length of hospital stay, length of follow-up, symptoms, need for redo FP). Short-term outcome was defined as the rate of surgical complications within the first 30 postoperative days and the long-term outcome was defined as the rate of Re-FP up to the latest available postoperative data.

Data analysis

Statistical analyses were performed using SPSS Statistic Version 25 (IBM Statistics, Armonk, NY, USA) for Windows. All data generated or analyzed during this study are included in this article. Categorical variables are summarized as frequencies and percentages and were compared using a two-tailed Fisher’s exact test. Continuous variables are expressed as mean ± standard deviation and were compared by a two-tailed paired t test. Odds ratios with 95% confidence intervals were calculated to determine the impact on patient outcome for each recorded variable. A value of p < 0.05 was considered statistically significant for all analyses.

Results

Overall group

A total of 302 children and young adults underwent FP during the study period. Of these, 128 (38.9%) cases were excluded due to the following criteria: lost to follow-up (n = 64, 19.5%), available data of follow-up less than 6 months (n = 55, 16.7%), incomplete data sets (n = 5, 1.5%), and previous organ transplantation (n = 4, 1.2%). The study group included 182 patients who underwent a total of 201 FP procedures.

Intraoperative complications included bleeding from a stomach vessel (n = 3), which could not be resolved laparoscopically and required a conversion to an open suture in two cases. Postoperative surgical complications leading to a reintervention within the first 30 postoperative days following FP included a laparoscopic release of a tight hiatoplasty suture as a cause of early and persistent postoperative dysphagia (n = 1), relaparotomy for secondary gastric perforation (n = 3) and leakage of a gastrostomy (n = 1), relaparotomy for dislocation of a gastrostomy tube with peritonitis (n = 1), and drainage of an abdominal abscess (n = 1).

In 24 cases (11.9%) a re-FP was necessary for recurrent and medically refractory GERD within the follow-up period of 53.4 ± 44.5 months. During this follow-up period, there was no operation-related mortality, although four neurologically impaired children (2.0%) died due to cardiorespiratory failure unrelated to surgery.

Effect of a previous abdominal surgery

Overall, 95 (47.3%) of the FP procedures were performed after a PAS. The previous operations are described in Table 1. The subgroup analysis results between patients with and without PAS at the time of FP are presented in Table 2.

Table 1 Previous abdominal surgery at time-point of fundoplication (multiple possible)
Full size table
Table 2 Subgroup analysis of patients with previous abdominal surgery (PAS) and without previous abdominal surgery (control) at time of fundoplication
Full size table

A PAS at the time of FP was associated with an odds ratio of 1.808 (95% CI 0.161–20.260, p = 0.631) for intraoperative complications and an odds ratio of 1.933 (95% CI 0.787–4.748; p = 0.150) for a redo FP.

Gastrostomy was performed in one third of cases before FP (67/201 = 33.3%). These patients showed comparable surgery times at FP (148.4 ± 47.5 min. vs. 134.3 ± 60.4 min; p = 0.069) and intraoperative complication rates (1/67 = 1.5% vs. 2/134 = 1.5%; p = 1.000). The length of hospital stay was also significantly longer for patients with gastrostomy (10.9 ± 7.7 days vs. 7.2 ± 3.9 days; p < 0.001) than for patients without previous gastrostomy. The need for redo FP was comparable between both groups (6/67 = 9.0% with gastrostomy vs. 18/34 = 53.0% without gastrostomy, p = 0.490) and a gastrostomy at the time of FP was associated with an odds ratio of 0.634 (95% CI 0.239–1.680, p = 0.359) for the need of a redo FP.

Patients with FP as a PAS showed comparable operation times (156.6 ± 54.6 min. vs. 136.2 ± 56.7 min.; p = 0.088) and intraoperative complication rates (1/27 = 3.7% vs. 2/174 = 12%, p = 0.523). The length of hospital stay was significantly longer (6.9 ± 3.0 days vs. 8.6 ± 6.0 days; p = 0.154) compared to patients without previous FP.

Fundoplication as PAS at the time of FP was associated with an odds ratio of 3.308 (95% CI 0.290–37.784, p = 0.336) for intraoperative complications and an odds ratio of 0.553 (95% CI 0.122–2.497; p = 0.553) for the need for redo FP.

Discussion

The aim of the present study was to analyze the impact of PAS on the FP-related outcome in a cohort of children and young adults undergoing FP for medically refractory GERD. We showed that PAS was associated with a significantly longer operation time and hospital stay, and, as the rate of intraoperative complications and the rate of redo FP were not significantly different, the presence of PAS did not have a significant influence on the overall outcome of the patients.

This is accordance with the literature regarding the safety of performing FP in children with PAS, such as gastrostomy or ventriculo-peritoneal shunts [15, 36], previous FP [10, 33], or other laparotomies [8, 29].

The clearly extended time of operation is most likely due to adhesions caused by the PAS, which may lead to suboptimal intraoperative line of sight and a technically more demanding operation [11, 12]. According to the literature, in our study PAS was associated with a higher conversion rate during FP (0–11%; [6, 8, 11, 12, 19,20,21,22,23]). Although the operation time was longer and the conversion rate was higher in children with PAS compared to children without PAS, both groups showed comparable intraoperative complication rates. These results are almost identical to results in the current literature since the overall intraoperative complication rates are reported as 0.8–14% [11, 12, 19,20,21, 24, 25] and as 0–17% in children with PAS [10, 14,15,16,17, 22, 26].

The laparoscopic approach for FP is associated with a shorter hospital stay [11, 27, 28]. This possibly explains the longer hospital stay in children with PAS in the present study, as in these patients the open FP rate was significantly higher than for children without PAS. Moreover, the longer hospital stay in children with PAS could also be partly due to the higher rate of neurologically impaired patients in this group, since according to the literature a significant comorbidity might prolong the hospital stay [11].

In present study, the redo FP rate for patients and young adults with PAS was comparable to the rate reported in the literature (0–18%; [8, 10, 29,30,31]). Although the regression analysis showed a tendency for a higher rate of redo FP in children with PAS, this was not statistically significant.

In some studies, the presence of a gastrostomy is mentioned as a major factor for the occurrence of postoperative complications following a pediatric FP [21, 25]. In our analysis, the postoperative complication rate was not influenced by the presence of PAS; however, the length of hospitalization was significantly longer in children with PAS and gastrostomy. In patients with a neurological impairment, gastrostomy was found significantly more frequently in our study, and therefore the presence of such a comorbidity could be a contributory cause to the prolonged hospital stay [11].

The presence of a previous FP as PAS did not impact the complications rate, the operation time, or the need for redo FP in the present study, while the rates of these parameters were comparable to published data [9, 10, 30, 32,33,34,35].

This study is limited by the single-center analysis, which limits the size of the study population and the retrospective nature of the study, leading to potential bias. A considerably higher number of cases for accurate evaluation might be accomplished through the widespread use of a centralized register. Moreover, a prospective, blind, randomized, two-arm study could avoid this bias and determine the procedure-related impact with more accuracy. However, such a study is probably unfeasible for ethical reasons; therefore, retrospective studies like ours currently represent one possible source of data on which to base recommendations in the clinical setting.

Conclusion

In children and young patients with a previous abdominal surgery, fundoplication can be performed safely for medically refractory gastroesophageal reflux disease. A previous abdominal surgery does not represent a contraindication or a limitation to performing fundoplication on these patients.