Annals of Otology, Rhinology d Laryngology 121(3):145-15O.© 2012 Annals Publishing Company. All rights reserved.

Variability of Postoperative Esophagrams After EndoscopieCrieopharyngeal Myotomy: Technique Dependence

Craig E. Berzofsky, MD; Roy A. Holiday, MD; Michael J. Pitman, MD

Objectives: We illu.strate the dependence of postoperative day (POD) 1 esophagram findings on the closure techniqueused after endoscopie crieopharyngeal myotomy (ECPM).Methods: We performed a retrospective chart review of POD 1 fluoroscopic examinations of the cervical esophagus uti-lizing contrast dye after ECPM to assess radiologie findings associated with three different techniques of addressing theexposed bueeopharyngeal fascia (BPF).Results: Each technique resulted in specific and different findings on the POD 1 esophagram. When the BPF was untreat-ed, the esophagram demonstrated a pseudodiverticulum with free flow of contrast dye. When a fibrin glue seal was used,the esophagram demonstrated a curvilinear focus of contrast dye projected over the retropharyngeal soft tissue persistingafter the swallow, similar to a leak. When fibrin glue application was combined with single-suture reapproximation of themucosal incision, the pattern was similar to esophagrams performed 6 weeks after myotomy.Conclusions: Different techniques used to address the exposed BPF following ECPM result in specific findings on thePOD 1 esophagram. Recognition of these imaging differences and open communication with the fluoroscopist will avoida misdiagnosis of a pharyngeal leak, which might cause an unnecessary delay of oral feeding and hospital discharge.Key Words: crieopharyngeal achalasia, edeopharyngeal dysfunction, dysphagia, endoscopie cricopharyngeal myotomy,radiology.

INTRODUCTION

Endoscopie cricopharyngeal myotomy (ECPM)has become an accepted surgical treatment for pa-tients with dysphagia associated with cricopharyn-geal muscle (CPM) dysfunction. Multiple techniquesare available to address the exposed bueeopharyn-geal faseia (BPF) at the completion of the surgery.The postoperative eare protocols mentioned in theliterature generally inelude a postoperative esopha-gram, whieh direets aeute eare. This study examinesthe relationship between the elosure technique usedand the findings on the postoperative day (POD) 1esophagram and examines the impaet of this rela-tionship on patient management.

Cricopharyngeal Dysfiinction and Its Morbidity.The CPM attaehes to the dorsal lateral portion ofthe erieoid eattilage bilaterally. The eomposition ofthe muselé is of fast- and slow-twiteh striated skele-tal muselés. This eomposition allows basal tone andfast aetion for swallowing or eructation. The CPMforms the upper esophageal sphineter, in conjunc-tion with the erieoid and arytenoid eartilages andthe inferior eonstrietor muselés.' Dysfunetion of the

CPM ean lead to dysphagia via obstruetion of theupper esophageal sphineter seeondary to hypertro-phy or aehalasia. The morbidity of this proeess is ona speetrum from minimal diff ieulty in handling solidfoods to severe dysphagia that results in signifieantweight loss and malnutrition.

Surgical Technique. The definitive treatment forerieopharyngeal dysfunetion is via a ericopharyn-geal myotomy. Kaplan- first deseribed the open ap-proach to this surgery in 1951 for the treatment ofdysphagia in a patient who had had poliomyelitis.In 1994, ECPM with a potassium titanyl phosphate(KTP) laser was first deseribed. Soon afterward, theearbon dioxide (C02) laser was employed, and ithas beeome the laser of ehoiee beeause of its abilityto eoagulate small vessels and minimise the unin-tended spread of thermal damage.^

Endoseopie erieopharyngeal myotomy is per-formed under general anesthesia. The teehnique hasbeen extensively deseribed elsewhere,"* but a briefsynopsis is neeessary here. A divertieuloseope isused to visualize and expose the CPM. The CPM isidentified as a mound bulging at the esophageal in-

From the Voice and Swallowing Institute, Department of Otolaryngology (Berzofsky, Pitman), and the Department of Radiology (Holi-day), New York Eye and Ear Infirmary, New York, New York.Correspondence: Michael J. Pitman, MD, Voice and Swallowing Institute, Dept of Otolaryngology, New York Eye and Ear Infirmary,310 E 14th St. 6th Floor, New York. NY 10003.

145

146 Berzofsky et al, Esophagram After Cricopharyngeal Myotomy

Fig 1. Intraoperative photographs. A) Endoscopie ericopharyngeal myotomy (ECPM) with exposed buccopharyngealfascia (BPF; dashed arrow) left to granulate. B) Single-suture closure (solid arrow) with retromucosal pocket filled withfibrin glue (dashed arrow).

troitus. A CO2 laser is used to incise the CPM in themidline. Each fiber of the CPM is transected untilthe underlying BPF is exposed. The BPF is formedby the visceral layer of the middle layer of deep cer-vical fascia. It forms an anatomic barrier that sealsoff the retropharyngeal space, which is posterior tothe fascia. This space is filled with areolar tissue andis bordered posteriorly by the alar fascia, which pro-tects the danger space. Violation of the BPF and thealar fascia and spaces posterior to it will increase therisk of postoperative mediastinitis. Although medi-astinitis has not been reported after ECPM, it is thisanatomy and the risk of mediastinitis that concernssurgeons. The lack of mediastinitis may be due tothe fact that the areolar tissue of the retropharyn-geal space remains as a barrier against diffusion ofpharyngeal contents, whereas in the transcervicalapproach this barrier is completely violated and itsprotection is eliminated.

Once complete transection of the CPM isachieved, closure of the surgical bed is an option. Ifthe BPF is hardy and intact, then closure of the areais not required and the area can safely granulate, asin Fig lA. If the BPF is violated or thin, it shouldbe addressed. The most common method is cover-ing the open surgical bed with fibrin glue (Baxter,Westlake Village, Califomia).^ A third, previouslyundescribed method includes placing one horizontalsuture in the esophageal mucosa to create a retro-mucosal pocket. A 5-0 Vicryl suture on a P-2 needle(Ethicon, Somerville, New Jersey) is used with theWeerda Diverticuloscope (Karl Storz, Tuttlingen,Germany). In using the Slimline Diverticuloscope(Karl Storz), a 5-0 plain gut suture (Ethicon) on an

S-14 needle is recommended, because it is easier tohandle in the smaller-lumen scope. The fibrin glueis then placed in this pocket. The advantage of thepocket is that it holds and concentrates the fibringlue over the BPF and incorporates the esophagealmucosa as another barrier. Upon inspection of theretromucosal pocket with an endoscope, it appearsthat this suture does not place enough tension on themucosa to medialize the underlying CPM, riskingreestablishment of the CPM and its dysfunction. Inaddition, other studies suggest that the placementof fibrin may prevent tissue adhesion, extrapolatedin this case to the ends of the transected CPM (FigIB).

The benefits of fibrin glue are multifactorial. Fi-brin glue acts as a sealant for the exposed and pos-sibly violated BPF to prevent extravasation throughthe surgical defect. This glue has been utilized in-stead of surgical repair in other fields, to preventpancreatic and postoperative fistulas.^-^ Also, fibringlue has been shown to prevent postoperative adhe-sions, most notably in endometriosis surgery.*^ Caremust be taken with the use of fibrin glue. There havebeen several case reports of anaphylactic reactionsto the aprotinin molecule used in the solution, andthere have been reported cases of parvovirus B19transmission with fibrin sealant, although no trans-missions of human immunodeficiency virus havebeen reported. Because of the risk of allergy to apro-tinin, the patient must be warned about this possibleadverse effect.^"'̂ Depending on the closure method,the findings on the POD 1 esophagram will vary. Ifthese variations are not recognized as changes con-sistent with alternative closure techniques, they may

Berzofsky et al, Esophagram Afler Cricopharyngeal Myotomy 147

be misinterpreted and result in an unnecessary alter-ation in care and increased patient morbidity.

Postoperative Feeding. The resumption of feed-ing after ECPM varies according to institution. Takeset al'^ obtain chest and lateral neck radiographs onPOD I. If free air is not seen, and no ciinicai signsor symptoms of infection are noted, then oral feed-ing begins on POD 1. If free air is identified, thenthe patient begins feeding on POD 3.'-* Lawsonand Remade'' perform chest radiographs on POD1. Their patients are fed parenterally for 3 days andthen begin a liquid and semisolid diet.'' Lim'"* per-forms an esophagram on POD 0 and begins feedingif the study is negative for a pharyngeal leak. At ourinstitution, we perform an esophagram on POD 1. Ifit is negative, the patient starts a clear liquid diet. Ifa leak is suspected, the patient will be kept on NPO(nil per os) status, and a follow-up esophagram willbe performed in 48 hours. Postoperative use of anesophagram to screen for a pharyngeal leak is oneof the most common elements of postoperative careafter ECPM.

In caring for these patients at our institution, wenoted that the POD 1 esophagram findings var-ied considerably, depending on the closure tech-nique used. At times, the findings were equivocalfor a pharyngeal leak. In 1 case, a patient was kepton NPO status, and discharge was delayed when apharyngeal leak was suspected from the esopha-gram findings, but in retrospect it was found thata leak had not occurred. What had been observedwere changes secondary to the closure technique.As such, it is prudent to understand and recognize

the differences in esophagram findings based on theclosure technique. The purpose of this investigationwas to review how POD 1 esophagrams after ECPMare affected by the closure technique and to reviewthe importance of these effects in clinical decision-making and patient care.

MATERIALS AND METHODSInstitutional Review Board approval was obtained

according to the protocol for retrospective chart re-views at New York Eye and Ear Infirmary. We re-viewed the operative reports of 17 patients who un-derwent ECPM and their postoperative esophagramsto elucidate findings consistent with three differentclosure techniques. All patients had a diagnosis ofcricopharyngeal dysfunction and underwent ECPMwith a C02 laser. The three closure techniques uti-lized were 1) no closure, 2) fibrin glue placed overthe BPE, and 3) single-suture reapproximation ofthe esophageal mucosa with a retromucosal pocketfilled with fibrin glue.

All patients were kept on NPO status overnight,and we obtained a Gastrografin (Braceo Diagnos-tics, Princeton, New Jersey) contrast dye esopha-gram on POD 1. The patients received a test dose ofno more than 5 mL of contrast dye before the study.Four subsequent swallows of 10 to 15 mL werethen performed and evaluated in the right anterioroblique, anterior-posterior, lateral, and left anterioroblique planes. Ifaspiration was noted at any time,the examination was terminated. Gastrografin waschosen for these examinations because of the con-cern for a pharyngeal leak. Gastrografin has a lower

Fig 2. Postoperative day (POD) 1 esophagrams of 2 patients with BPF left to granulate. Pseudodiverticulum (outlined byarrows) is contained, despite retropharyngeal emphysema.

148 Berzofsky et al, Esophagram After Cricopharyngeal Myotomy

Fig 3. POD 1 esophagram.s of 2 patients with fibrin glue placed on exposed BPF. There is retained contrast dye in thick-ened retropharyngeal soft tissue at C6-7 (outlined by arrows).

incidence of mediastinitis than does barium in thissituation. Although there is a risk of pulmonary ede-ma and pneumonitis if copious amounts of contrastdye are aspirated, aspiration is minimized with theabove precautions, and all patients were monitoredin a hospital. A clear liquid diet was initiated in theabsence of a pharyngeal leak as determined by thesurgeon and the radiologist.

RESULTSThree different esophagram patterns were noted

on POD 1. No esophagram revealed extravasationof contrast dye material. The first pattern is a pseu-dodiverticulum without extravasation and an ab-sence of retained contrast dye in the pseudodiver-ticulum (Fig 2). This pattern occurred only with thetechnique of using no closure and allowing the BPFto granulate. This pattern occurred in both patientswith this closure technique.

The second pattern is a pseudodiverticulum withretained contrast dye in the surgical bed, which doesnot clear with repeated swallows. This occurs after

Fig 4. POD 1 esophagrams of 2 patients after placement of one suture and fibrin glue in retromucosal space. Esopha-grams are similar to those at 6 weeks, as demonstrated in Fig 5.

Berzofsky et al, Esophagram After Cricopharyngeal Myotomy 149

closure with only fibrin glue placement. The con-trast dye was curvilinear in nature at the site of thesurgical bed (Fig 3). The first patient with this find-ing was kept on NPO status because of the high like-lihood of a pharyngeal leak. A scout film on POD2 revealed a resolution of the contrast dye residuefrom the previous day. The repeat esophagram thenshowed the same curvilinear contrast dye residue asseen on POD 1. It was then assumed that the curvi-linear focus was not a leak, but the result of the clo-sure technique. This finding was consistently notedin all 9 patients who subsequently had this closuretechnique and an esophagram on POD 1.

The third finding was a normal-appearing esopha-gram. This was seen in patients with fibrin glueplacement in a retromucosal pocket (Fig 4). Thesepatients showed no signs of retained contrast dye orpseudodiverticulum, and the esophagram was simi-lar to those performed after complete healing at 6weeks independent of closure technique. This find-ing occurred in all 4 of the patients with this closuretechnique.

DISCUSSION

The most devastating outcome of ECPM wouldbe mediastinitis from a pharyngeal leak. The leakcould occur by extravasation through the surgicaldefect in the BPF. Although there have been no re-ports of this complication, surgeons have utilizednumerous protocols as discussed above to monitorpatients after surgery. The method used at this insti-tution is to obtain a POD 1 esophagram to examinethe integrity of the surgical defect. After reviewingthe experience at this institution, we realized thatthree different results were obtained on postopera-tive esophagrams. These results were found to bedependent on the closure technique.

The pseudodiverticulum seen in the first closuretechnique is the void created by lateralization ofthe CPM. There was no concern of a leak in thisinstance, as all of the contrast dye was cleared onthe initial swallow. All patients granulated over theBPF, and the 6-week postoperative esophagrams ap-peared normal (Fig 5).

Initially, after closure with only fibrin glue placedover the BPF, there was a concern by the surgeonand the radiologist for a leak due to the persistenceof contrast dye after the completion of the swallow.The patient was kept on NPO status for another dayand had a repeat esophagram on POD 2. The scoutfilm for this patient showed resolution of the re-tained contrast dye, but repeat examination repro-duced the original findings of a curvilinear focus of

Fig 5. Six-week postoperative esophagram of patient inFig 2. Note resolution of pseudodiverticulum seen onPOD 1 esophagram.

contrast dye.

It was hypothesized that the linear area of ques-tionable contrast dye extravasation was the site atwhich fibrin glue was placed. Although no reportswere found discussing the interaction of contrastdye and fibrin glue, it was postulated that contrastdye could be coating and adhering to the fibrin glue.The time between examinations allowed for dissipa-tion of the contrast dye material, which would nothave occurred if it was intraparenchymal. The con-clusion that the linear residue of contrast dye didnot represent extravasation was confirmed, as thispattern was consistently observed in 9 subsequentpatients who had a similar closure technique and aPOD 1 esophagram.

Last, in patients with a single suture placed to cre-ate a retromucosal pocket filled with fibrin glue, thePOD I esophagram appeared normal and similar tothose performed 6 weeks after the operation. Whencompared to full closure of the mucosa with mul-tiple sutures, this technique is more rapid, is morewatertight and airtight, and provides an extra layerof closure. This technique should decrease the riskof mediastinitis, as well as pneumomediastinum andsubcutaneous emphysema. Anecdotally, the patientsappear to have less discomfort than those who un-dergo the other closure techniques. Although it hasnot been proven, on visual inspection this techniquedoes not appear to result in medialization of theCPM. In addition, the fibrin glue may minimize therisk of adherence between the ends of the severedCPM, further decreasing the risk of its reconstitu-

150 Berzofsky et al, Esophagram Afler Cricopharyngeal Myotomy

tion and symptom reeurrence. This is the standardelosure method now used at our institution in pa-tients who undergo ECPM.

Considering the absence of evidence of a pharyn-geal leak of air or contrast dye in the 5 patients whohad the multilayer elosure teehnique, and the nor-mal appearance of the POD 1 esophagram, we havebecome confident in the integrity of this elosure.Thus, we now begin patients on a elear liquid dieton the evening of POD 0 and do not obtain a POD1 esophagram. Patients are discharged at midday onPOD 1 if their condition is clinically stable.

CONCLUSIONSThe elosure technique in ECPM consistently af-

fects the appearanee of the POD I esophagram. Forsurgeons who are beginning to perform ECPM, er-roneous findings of extravasation may result in pre-mature abandonment of this benefieial procedurebeeause of the false presumption that it resulted in apharyngeal leak. Reeognition of the effeets that dif-ferent closure teehniques have on the postoperativeesophagram is essential to optimizing patient eare.Our protoeol has ehanged beeause of our evaluationand understanding of the information that these ra-diographs have provided.

REFERENCES1. Sivarao DV, Goyal RK. Functional anatomy and physiol-

ogy of the upper esophageal sphineter. Am J Med 2000; 108(sup-pl4a):27S-37S.

2. Kaplan S. Paralysis of deglutition, a post-poliomyelitiscomplication treated by section of the cricopharyngeus muscle.AnnSurg l951;l33:572-3.

3. Herberhold C, Walther EK. Endoseopie laser myotomy inerieopharyngeal achalasia. Adv Otorhinolaryngol 1995;49:144-7.

4. Pitman M, Weissbrod P. Endoseopie CO2 laser cricopha-ryngeal myotomy. Laryngoscope 2009; 119:45-53.

5. Lawson G, Remade M. Endoseopie cricopharyngealmyotomy: indications and technique. Curr Opin OtolaryngolHead Neck Surg 2006; 14:437-4L

6. Choi YY, Cho JY, Kim YJ. Successful endoseopie man-agement of an esophagojejunostomy leak using fibrin glue in-jection after a total gastreetomy. Am Surg 2011 ;77:376-8.

7. Rábago L, Ventosa N. Castro JL, Marco J, Herrera N.Gea F. Endoseopie treatment of postoperative fistulas resistantto eon.servative management using biologieal fibrin glue. En-doseopy 2002;34;632-8.

8. Takeuchi H, Awaji M, Hashimoto M, Nakano Y, Mitsu-hashi N, Kuwabara Y. Reduction of adhesions with fibrin glueafter laparoscopic excision of large ovarian endometriomas. JAm Assoc Gynecol Laparosc 1996;3:575-9.

9. Kanazawa R. Sato S, Iwamoto N, Teramoto A. Allergicreaction following arachnoid plasty with a fibrin sealant. NeurolMed Chir (Tokyo) 2010;50:608-10.

10. Schievink Wl, Georganos SA, Maya MM. Moser FG,Bladyka M. Anaphylactie reactions to fibrin sealant injectionfor spontaneous spinal CSF leaks. Neurology 2008;70:885-7.

11. Kawamura M. Sawafuji M. Watanabe M, Horinouchi H,Kobayashi K. Frequency of transmission of human parvovirusBI9 infection by fibrin sealant used during thoracic surgery.Ann Thorac Surg 2002;73:1098-100.

12. Joeh C. The safety of fibrin sealants. Cardiovasc Surg2003;ll(suppl l):23-8.

13. Takes RP, van den Hoogen FJ, Marres HA. Endoseopiemyotomy of the crieopharyngeal muscle with CO2 laser sur-gery. Head Neek 2005;27:703-9.

14. Lim RY. Endoseopie CO2 laser cricopharyngeal myoto-my. J Clin Laser Med Surg 1995;13:241-7.

Copyright of Annals of Otology, Rhinology & Laryngology is the property of Annals Publishing Company andits content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder'sexpress written permission. However, users may print, download, or email articles for individual use.