ORIGINAL CONTRIBUTIONS

Esophageal Dysmotility After Laparoscopic GastricBand Surgery

Philip A. Le Page & Sebastianus Kwon &

Sarah J. Lord & Reginald V. Lord

# Springer Science+Business Media New York 2013

AbstractBackground The effect of the laparoscopic adjustable gastricband (LAGB) on the esophagus has been the subject of fewstudies despite recognition of its clinical importance. The aimof this study was to investigate the frequency and clinicaleffect of esophageal dysmotility and dilatation after LAGB.Methods We undertook a retrospective analysis of 50 consec-utive patients with no dysmotility on perioperative videocontrast swallow who underwent primary LAGB operation.All patients had serial focused postoperative contrast studiesfor band adjustments at least 6 months post-LAGB. Clinicaland radiological outcomes were assessed.

Results Median follow-up time was 18 months (range 7–39 months), and the median number of contrast swallowsper patient was 5. The mean excess weight loss (EWL) overallwas 47 % (standard deviation (SD) 22.3). Radiological abnor-malities were recorded in 17 patients (34 %, 95 % confidenceinterval (CI) 21–49 %), of whom 15 had radiologicaldysmotility and 7 had esophageal dilatation (five patientshad both dysmotility and dilatation). Of these 17 patients,six (35 %) developed significant symptoms of dysphagia,gastroesophageal reflux disease (GERD) or regurgitation re-quiring fluid removal. In comparison, 12 of 33 (36 %) patientswithout radiological abnormalities developed symptoms re-quiring fluid removal (p =1.00). Patients with radiologicalabnormalities were significantly older than those without the-se abnormalities. Symptoms were alleviated by removingfluid in most patients.Conclusions The LAGB operation results in the developmentof radiological esophageal dysmotility in a significant propor-tion of patients. It is not clear if these changes are associatedwith an increased risk of significant symptoms. Fluid removalcan reverse these abnormalities and their associatedsymptoms.

Keywords Obesity . Morbid obesity . Gastric band .

Laparoscopic gastric band . Bariatric surgery . Esophagealdysmotility . Gastroesophageal reflux

Introduction

Laparoscopic adjustable gastric band (LAGB) placement is acommon bariatric operation that provides significantly betterweight loss and reduction in obesity-related comorbid ill-nesses, especially type 2 diabetes mellitus [1–4], comparedto conservative treatment options [1, 5–7].

P. A. Le Page : R. V. LordDepartment of Upper Gastrointestinal Surgery, St. Vincent’sHospital, Victoria St, Darlinghurst, Sydney, NSW 2010, Australia

S. Kwon :R. V. LordGastroesophageal Cancer Research Program, St. Vincent’s Centre forApplied Medical Research, Darlinghurst, Sydney, Australia

S. Kwone-mail: sebkwon@gmail.com

S. J. LordDepartment of Epidemiology and Medical Statistics, School ofMedicine, University of Notre Dame Australia, Darlinghurst,Sydney, NSW 2010, Australiae-mail: sally.lord@nd.edu.au

R. V. LordDepartment of Surgery, School of Medicine, University of NotreDame Australia, Darlinghurst, Sydney, NSW 2010, Australia

P. A. Le Page (*) : R. V. Lord (*)St Vincent’s Clinic, Suite 606, 438 Victoria Street, Darlinghurst,Sydney, NSW 2010, Australiae-mail: phillepage@yahoo.com.aue-mail: recep606@stvincents.com.au

R. V. Lorde-mail: rvlord@stvincents.com.au

OBES SURGDOI 10.1007/s11695-013-1134-5

The advantages of the LAGB operation are the low risk ofmajor complications compared with the other bariatric opera-tions and its reversibility [8, 9]. Disadvantages include theneed for postoperative band adjustments, the occurrence ofobstructive episodes during eating, and the need for bandremoval in a significant proportion of patients due to slippage,erosion, or difficulties with compliance and tolerability[7, 10].

The adverse effects of LAGB on the esophagus are increas-ingly well recognized. Although LAGB reportedly provides asignificant improvement in gastroesophageal reflux as mea-sured by distal esophageal acid exposure [11], the band cancause a bland regurgitation of recently ingested foods anddrinks from the gastric pouch, which may not be detected bypH monitoring [11–13]. This reflux may manifest as cough,including nocturnal cough, as well as the typical reflux symp-toms of heartburn and regurgitation. Post-LAGB reflux symp-toms are variably improved by acid suppressant medicationssuch as PPIs but usually need fluid or band removal forcomplete resolution.

Esophageal dysmotility and dilatation as complications ofLAGB have been reported in up to 69 and 26 % of patients,respectively [14–17]. These complications are thought to re-sult in reduced esophageal clearance, which also increases theseverity of gastroesophageal reflux [14, 18, 19]. Esophagealmanometry studies performed at our institution on patientsreferred from other centers demonstrated that changes ofsevere dysmotility can be present after LAGB. These patientsdid not have preoperative motility testing, and their postoper-ative symptoms were severe enough to warrant referral formotility studies, so the significance of these observations isuncertain. It, nevertheless, seems noteworthy that 11 of the 13patients studied by manometry at our institution had esopha-geal dysmotility, which was, primarily, low-amplitude bodycontractions with disordered peristalsis includingpseudoachalasia patterns (median amplitude 45 mmHg, range0–125 mmHg) or hypotonic lower esophageal sphincter. The-se findings together with the substantial rate of regurgitationand cough in the patients who underwent LAGB at thisinstitution prompted this novel study on the radiologicalchanges in the esophagus induced by LAGB and the clinicaleffects of these changes.

Methods

Patients

Fifty consecutive patients who underwent LAGB operationperformed by the senior author (RVL) at this institution werestudied retrospectively if they met the following inclusioncriteria: (1) no history of previous gastroesophageal surgery;(2) no evidence of esophageal dysmotility, dilatation, or other

esophageal abnormality apart from hiatus hernia on a periop-erative contrast study (preoperative in 47 patients and firstpostoperative day in three patients); and (3) at least onepostoperative contrast study, with clinical data, performedmore than 6 months after LAGB surgery. The swallow studieswere performed as part of routine clinical care. The objectiveof the preoperative X-ray was to detect severe dysmotility thatmight contraindicate the placement of the LAGB prosthesis orto detect a hiatus hernia needing repair at the time of LAGBplacement. The objective of the postoperative contrast swal-low studies was to accurately fill the band to an optimumvolume without the risk of port cannulation complications,creating a 3–4-mm luminal diameter with (ideally) few ad-justments. Approval to conduct the study as a low-risk studynot requiring individual patient consent was obtained from theethics committee for this institution.

The control perioperative contrast swallow method wasbased on the videoesophagram protocol developed at theUniversity of Southern California [20]. Essentially, prone,supine, and erect video recordings of the esophagus weretaken by asking the patient to swallow 10 ml boluses of mixedbarium contrast agent [20].

Operation

The patient was placed in modified Lloyd–Davies positionwith the surgeon between the patient's legs and the assistant tothe left of patient. Optical entry was obtained using a 12-mmEndopath Xcel Bladeless Trocar (Ethicon Endo-Surgery, NJ,USA). After creating the pneumoperitoneum, the Nathansonliver retractor (Cook Surgical Bloomington, Indianapolis,USA) was placed through a subxiphoid incision, and threeports were placed in the right and left upper quadrants and theleft flank. After excising the gastroesophageal fat pad andrepairing a hiatus hernia if present, a retrogastric path for theband was created under direct vision using the pars flaccidatechnique. A Swedish adjustable gastric band (Obtech,Ethicon Endo-Surgery, New Jersey, USA) was secured inplace with 0 Ethibond (Ethicon Endo-Surgery, NJ, USA)gastrogastric sutures. The port was subsequently secured onthe external oblique aponeurosis.

Clinical Follow-up

Patients were seen by the surgeon postoperatively on day 1and discharged on a bariatric fluid diet. They were seen againat 2 weeks, and the first band adjustment was performed at 3–6 weeks. The band adjustments were performed under fluo-roscopy in order to optimize cannulation and accurately adjustthe fluid in the band. Further adjustments were performedwhen clinically indicated.

OBES SURG

Radiological Band Adjustments and Assessmentof Esophageal Motility and Diameter

The X-ray was done by one of three senior attending radiol-ogists experienced in the technique. A screening shot wasinitially taken, and the port was cannulated with the patientin a 45° head-up position. The patient was asked to swallowbarium, and dynamic images were taken focusing around thelower esophagus, hiatus, and gastric pouch. Fluid was addedor removed depending on the surgeon's request and on theradiologists' assessment of contrast hold up in the pouch,stoma size, and the presence of esophageal dysmotility ordilatation.

Dysmotility was defined as being present when there wasclearly a significant abnormality in esophageal body motility,including the finding of stasis of contrast in the esophagealbody sufficient to obscure the mucosal folds or overlyingstructures, significant reverse peristalsis or incoordinated ter-tiary waves, marked delayed passage of contrast, or diffuseesophageal spasm-like findings. Dilatation was defined asbeing present when the diameter of the distal esophagus wasmarkedly increased when compared to the proximal esopha-gus or to that of the distal esophagus in a previous contrastvideoesophagram study. In the presence of dilatation ordysmotility, fluid was removed from the band until there weredynamic signs of radiological improvement.

Data Collection

Data were collected retrospectively by two upper gastrointes-tinal surgery fellows using the medical records, clinical patientfiles, radiology films, and reports. Variables recorded wereage, height, weight, and body mass index (BMI) at initialconsultation, date of procedure, dates and findings of contrastswallows and adjustments, follow-up duration, last BMI, sig-nificant symptoms of dysphagia or reflux, and need for fluidor band removal.

Outcomes of Interest

The primary outcomes studied were the proportion of patientswith radiological dysmotility and esophageal dilatation, asreported by the three consultant radiologists involved in thestudy. Dysmotility was defined as being present when therewas stasis of contrast in the esophagus sufficient to obscurethe mucosal folds or overlying structures, significant delay inthe passage of contrast through the esophagus, diffuse esoph-ageal spasm-like findings, or severe reverse peristalsis orincoordinated tertiary waves. Dilatation was deemed presentwhen there was a clear increased diameter distally whencompared to the proximal esophagus or to that of a previouscontrast swallow.

Symptoms that were routinely assessed and studied weredysphagia, reflux, or regurgitation requiring fluid removal.The effect of removing fluid from the band on these symptomswas also recorded and assessed.

Analysis

Continuous variables were summarized by calculating meanand standard deviation (SD) if normal distribution and medianand range if non-normal distribution. Categorical variableswere summarized by reporting frequencies. Statistical analy-ses were performed to explore associations between the pres-ence of postoperative radiological abnormalities and patientage and operation outcomes (percent excess weight loss(%EWL), symptoms, symptom resolution following fluid re-moval, and band removal). For these analyses, two-tailed ttests for independent samples were used for continuous vari-ables (mean age and %EWL), and Fishers exact chi-squaretest was used for categorical variables.For all statistical anal-yses, a p value of <0.05 was considered statistically signifi-cant. Statistical analyses were performed using SAS 9.3 (Cary,NC, USA).

Results

Patient Characteristics

Fifty consecutive patients meeting the entry criteria underwentLAGB between February 2008 and April 2011. The mean ageof these patients was 47 years (SD 11.4), and 72 % werefemale. The mean preoperative BMI was 41 (SD 7.2), andthe mean percentage of excess weight loss was 47 % (SD22.3). The median clinical follow-up was 18months (range 7–39). The median number of postoperative fluoroscopic studieswas 4 (range 1–10).

Risk of Esophageal Complications

During the study period, there was a 34 % risk (95 % confi-dence interval (CI) 21–49 %) of developing esophageal ab-normalities (Table 1), of which the most common was thedevelopment of dysmotility (15/50, 30 %). These were clas-sified as mild in six patients, moderate in seven patients, andsevere in two patients.

Esophageal dilatation was seen in seven (14 %) patients ata median of 16 months postoperatively (range 1–27). Five ofthese patients (71 %) also had dysmotility. The mean age ofpatients with radiological abnormalities was higher than thosewith no abnormalities (54 versus 43 years, p =0.001), whilebaseline BMI was similar for each group.

OBES SURG

Relationship to Symptoms and Effect of Fluid Removal

Patients both with and without radiologic abnormalities of theesophagus had similar rates of symptoms necessitating fluidremoval (36 versus 35 %, p =1.00, Table 1). Detailed follow-up data after fluid removal were available for 14 patients; ofthese, 11 (79 %) had symptom improvement, with similarresults for both groups (78 versus 80 %, p =1.00). Bothpatients who had esophageal dilatation alone improved withfluid removal, and four of the five patients with concurrentdysmotility improved after fluid removal.

The necessity for band removal, which may be a marker ofeven more significant problems, was investigated. More pa-tients with no abnormality required band removal (four versusone), but this was due to other problems such as slippage orerosion.

Focusing on symptomatic patients only, 18 of the total 50patients had symptoms; of whom, one third had a radiologicalabnormality (Fig. 1). Each of these patients had fluid removedin order to alleviate their symptoms. This provided symptomresolution in most patients, with symptoms persisting despitefluid removal in only three patients; of whom, one patient hada radiological abnormality present (Fig. 2).

Discussion

In this study, we assessed esophageal function before and afterLAGB placement by performing routine contrast video swal-low studies in 50 consecutive patients with a minimumfollow-up of 6 months. We found a 34 % risk of developmentof an esophageal abnormality at a median follow-up of18 months. A significant proportion developed dysmotility,and dilatation was also observed. We did not identify anystatistically significantly associated factors (%EWL, need for

fluid removal, symptom resolution after fluid/band removal)except that, interestingly, the average age of patients develop-ing dysmotility/dilatation was older than that of those withoutthese changes. This suggests that older patients may be moreprone to develop esophageal problems after LAGB placementand that consequently, the band should be filled more conser-vatively in this group.

Our findings are consistent with ranges found in otherstudies [14–17]. A higher prevalence of these problems wasreported by Naef et al. [15], who in a study of 167 patientswho underwent yearly barium swallow studies followingLAGB insertion, found a 68.8 % incidence of dysmotilityand a 25.5 % incidence of dilatation. Our lower rates mayrelate to the study design in that we excluded patients withpreoperative abnormalities by performing preoperative videostudies, whereas preoperative motility was not assessed in thestudy by Naef et al. Another factor that may explain thediffering results is that we used different criteria for diagnos-ing esophageal dilatation. We compared the distal and proxi-mal esophageal diameters in the same swallow study or thedistal diameter in previous studies from the patient, whereasNaef et al. measured the diameter of the esophagus. Addition-ally, our median follow-up was shorter than that reported by

Table 1 Characteristics and clinical outcomes for patients with and without radiological esophageal complications

All patients No abnormality Any abnormality P ª Type of abnormality

Dilatation Dysmotility

N (%) 50 33 (66 %) 17 (34 %) - 7 (14 %) 15 (30 %)

Age, mean (SD) 47 (11.4) 43 (10.5) 54 (10.1) 0.001 55.9 (9.9) 54.7 (11.2)

%EWL, mean (SD) 47 (22.3) 46.3 (24.2) 47.2 (17.8) 0.9 49.2 (17.7) 45.4 (16.9)

Fluid removal required for symptoms 18 (36 %) 12 (36 %) 6 (35 %) 1.00 2 (29 %) 6 (40 %)

Symptoms resolved after fluid removalb 11/14(79 %) 7/9 (78 %) 4/5 (80 %) 1.00 2/2 (100 %) 4/5(80 %)

Band removal required 5/50 (10 %) 4/33 (12 %) 1/17 (6 %) 0.7 1/7 (14 %) 0/15 (0 %)

N patient number, SD standard deviation, EWL excess weight loss

ªStatistical tests used were unpaired sample two-tailed t test to test for differences in means for age and %EWL for patients with radiological abnormalityversus those without abnormality and Fisher exact chi-square test for difference in proportions for patients with radiological abnormality versus thosewithout abnormalityb Denominator accounts only for patients with follow-up after fluid removal

Fig. 1 Proportion of symptomatic group with radiological abnormalities

OBES SURG

Naef et al., suggesting that dysmotility and dilatation maycontinue to develop over the longer term after LAGB.

A systematic review from 2010 [11] looked at studiescomparing pre- and postoperative manometry results, findingthat LAGB resulted in an increase in esophageal dysmotilityfrom 3.5 to 12.6 %. The overall prevalence was substantiallylower than that in our study. There is no clear explanation forthis lower rate of dysmotility. This is especially so whenconsidering that the comprehensive study by Fuller et al., inwhich 202 patients with foregut symptoms underwentvideoesophagram studies and also manometry of the Univer-sity of Southern California, indicated that the video study issignificantly less sensitive than manometric evaluation for thedetection of nonspecific esophageal motility disorders [20].Esophageal manometric testing is the most accurate methodfor diagnosing and classifying esophageal motility disorders,but it is also invasive and uncomfortable. By using the contrastswallow study to assess motility, patients were fully compli-ant, and this study could be conducted without needing anytests that were not part of routine clinical care.

We investigated the symptomatic significance of abnormalradiological findings. Significant symptoms were not moreprevalent in comparison to those who had no radiologicalabnormality, suggesting that dysphagia and reflux symptomsafter LAGB are mostly due to obstruction at the level of theband rather than an effect on the esophagus. In support of this,most (11 of 17) patients with dysmotility or dilatation did nothave significant symptoms needing fluid removal. Further-more, the proportion whose symptoms improved after fluidremoval was similar for each group. This suggests that theclinical significance of the contrast swallow abnormalitiesafter LAGB, despite beingmarked in some patients, is limited.For these reasons, we do not advocate performing fluoroscopyon all patients unless symptoms are severe or band adjust-ments are performed under fluoroscopy. Our radiology

department has provided reassurance that the radiation doseinvolved in adjusting under X-ray guidance is minimal be-cause of the targeted nature of the studies. It is also possible,however, that the apparently limited effect of esophagealdysmotility and dilatation are because the follow-up durationin this study was too short for the effects of abnormal esoph-ageal function to become manifest in many patients.

It is encouraging that esophageal abnormalities improvedsignificantly in the majority of patients by simply removingfluid from the band, as reported by other studies [19]. Thissuggests that despite the quite alarming findings on the videoswallow studies in some patients, these findings are reversibleif the band is decompressed promptly.

In conclusion, the LAGB operation results in the develop-ment of radiological esophageal dysmotility in a significantproportion of patients. Fluid removal can reverse these abnor-malities and their associated symptoms for the majority ofpatients.

Acknowledgments The authors thank Dr. Sebastian Fung, a radiolo-gist, for scoring the contrast swallow studies.

Conflict of Interest All authors declare no conflicts of interest.

References

1. Belachew M, Legrand MJ, Defechereux TH, et al. Laparoscopicadjustable silicone gastric banding in the treatment of morbid obesity.A preliminary report. Surg Endosc. 1994;8:1354–6.

2. O'Brien PE, McPhail T, Chaston TB, et al. Systematic review ofmedium-term weight loss after bariatric operations. Obes Surg.2006;16:1032–40.

3. Favretti F, Ashton D, Busetto L, et al. The gastric band: first-choiceprocedure for obesity surgery. World J Surg. 2009;33:2039–48.

4. Viardot A, Lord RV, Samaras K. The effects of weight loss andgastric banding on the innate and adaptive immune system in type2 diabetes and prediabetes. J Clin Endocrinol Metab. 2010;95:2845–50.

5. Sjostrom L, Narbro K, Sjostrom CD, et al. Effects of bariatric surgeryon mortality in Swedish obese subjects. N Engl J Med. 2007;357:741–52.

6. Adams TD, Gress RE, Smith SC, et al. Long-term mortality aftergastric bypass surgery. N Engl J Med. 2007;357:753–61.

7. BelachewM, Belva PH, Desaive C. Long-term results of laparoscop-ic adjustable gastric banding for the treatment of morbid obesity.Obes Surg. 2002;12:564–8.

8. Parikh MS, Laker S, Weiner M, et al. Objective comparison ofcomplications resulting from laparoscopic bariatric procedures. JAm Coll Surg. 2006;202:252–61.

9. Robinson MK. Surgical treatment of obesity—weighing the facts. NEngl J Med. 2009;361:520–1.

10. Victorzon M, Tolonen P. Intermediate results following laparoscopicadjustable gastric banding for morbid obesity. Dig Surg. 2002;19:354–7.

11. de Jong JR, Besselink MG, van Ramshorst B, et al. Effects ofadjustable gastric banding on gastroesophageal reflux and esophagealmotility: a systematic review. Obes Rev. 2010;11:297–305.

12. Crookes PF, Corkill S, DeMeester TR. Gastroesophageal reflux in

Fig. 2 Outcomes for symptomatic patients having fluid removed andproportions with radiological abnormality

OBES SURG

achalasia. When is reflux really reflux? Dig Dis Sci. 1997;42:1354–61.

13. Rebecchi F, Rocchietto S, Giaccone C, et al. Gastroesophageal refluxdisease and esophageal motility in morbidly obese patients submittedto laparoscopic adjustable silicone gastric banding or laparo-scopic vertical banded gastroplasty. Surg Endosc. 2011;25:795–803.

14. de Jong JR, Tiethof C, van Ramshorst B, et al. Esophageal dilationafter laparoscopic adjustable gastric banding: a more systematicapproach is needed. Surg Endosc. 2009;23:2802–8.

15. Naef M, Mouton WG, Naef U, et al. Esophageal dysmotility disor-ders after laparoscopic gastric banding—an underestimated compli-cation. Ann Surg. 2011;253:285–90.

16. O'Rourke RW, Deveney CW, McConnell DB, et al. Esophageal

motility disorders after gastric banding. Dis Esophagus. 2007;20:269–73.

17. Suter M, Dorta G, Giusti V, et al. Gastric banding interferes withesophageal motility and gastroesophageal reflux. Arch Surg.2005;140:639–43.

18. Lew JI, Daud A, DiGorgi MF, et al. Preoperative esophageal ma-nometry and outcome of laparoscopic adjustable silicone gastricbanding. Surg Endosc. 2006;20:1242–7.

19. Khan A, Ren-Fielding C, Traube M. Potentially reversiblepseudoachalasia after laparoscopic adjustable gastric banding. JClin Gastroenterol. 2011;45:775–9.

20. Fuller L, Huprich JE, Theisen J, et al. Abnormal esophageal bodyfunction: radiographic-manometric correlation. Am Surg. 1999;65:911–4.

OBES SURG