Postcricoid Cushion” Observations on the Vascular Anatomy of the Posterior Cricoid Region

ORIGINAL ARTICLE The “Postcricoid Cushion” Observations on the Vascular Anatomy of the Posterior Cricoid Region

Stephen R. Hoff, MD; Peter J. Koltai, MD

Objective: To describe the cyclical vascular enlarge- months had presence of a cushion compared with only ment that occurs in the postcricoid region during the ex- 38% of children 24 months or older (PϽ.001). Twenty- piratory phase on an infant’s cry, and to consider the ana- five percent of the cushions had violaceous discolor- tomic, physiologic, and clinical implications of this ation that resembled a vascular malformation. phenomenon, which we term the “postcricoid cushion.” Conclusions: Anatomic studies have demonstrated a rich Design: A total of 125 consecutive office fiber-optic la- venous plexus in the postcricoid region of the larynx. Dur- ryngoscopic examinations in children and infants were ing the expiratory phase of an infant’s cry, there is a cy- reviewed for engorgement and vascular discoloration of clical engorgement, occasionally with vascular discolor- the postcricoid region. Presence of a postcricoid cush- ation, in the postcricoid region at the same level of the ion in relation to patient age was reviewed. A compre- venous plexus—the “postcricoid cushion.” We propose hensive literature review was also performed. that during crying, with acute elevation in intrathoracic pressure, there is a filling of the plexus, causing apposi- Setting: Tertiary care pediatric hospital. tion of the postcricoid cushion against the posterior pharyngeal wall, which may serve as a protective barrier to Patients: Patients from newborns to 17 years old un- dergoing laryngoscopy for any reason. emesis in infants. Our observations relate and differen- tiate this normal physiologic phenomenon from the rare Results: Sixty-one percent of the videos showed a post- cases of postcricoid vascular anomalies. cricoid cushion with cyclical enlargement during crying. Eighty-eight percent of children younger than 24 Arch Otolaryngol Head Neck Surg. 2012;138(6):562-571

E BECAME AWARE nounced and most regularly observed in of the vascular newborns and infants. We term this en- anatomy of the largement the “postcricoid cushion.” postcricoid region by a cluster of 4 Videos available online at children seen from 1998 through 2002 W 1 www.archoto.com who had “postcricoid hemangiomas” (Figure 1). Informal analysis of find- Further insight into the morphologic ings from subsequent office fiber-optic la- significance of the postcricoid cushion ryngoscopies suggested that a violaceous occurred serendipitously during review bulge, which at times appeared much like Author Affiliations: Division of of a French anatomy text from 1854 by a postcricoid hemangioma, was more fre- Bourgery and Jacob12 in which there is Pediatric Otolaryngology, quently seen in this region than would be Children’s Memorial Hospital, a beautiful color illustration of the post- Chicago, Illinois (Dr Hoff); expected, given the rarity of these vascu- cricoid region, demonstrating a rich 2-11 Department of lar anomalies (Video 1; http://www vascular plexus at this site (Figure 2). Otolaryngology–Head and Neck .archoto.com). In looking for postcricoid Subsequent historical searches of the Surgery, Feinberg School of vascular coloration, we became con- anatomical literature confirmed that this Medicine, Northwestern scious of a consistent phenomenon not, plexus of veins has been extensively University, Chicago (Dr Hoff); to our knowledge, previously appreci- studied.13-24 The plexus is relatively more and Division of Pediatric ated: a cyclical engorgement of the post- prominent in fetal dissections than in Otolaryngology, Department of cricoid mucosa that was coincidental with Otolaryngology–Head and Neck adult ones. Surgery, Stanford University the expiratory phase of the infant’s cry Our hypothesis is that these observa- School of Medicine and Lucile (Video 2). Interestingly, the prominence tions are related and can be woven into a Packard Children’s Hospital, of these tissues, which occasionally has rational narrative that is anatomically, Stanford, California (Dr Koltai). vascular coloration, seems to be most pro- physiologically, and clinically consis-

ARCH OTOLARYNGOL HEAD NECK SURG/ VOL 138 (NO. 6), JUNE 2012 WWW.ARCHOTO.COM 562

D E F

Figure 1. Illustration of 3 cases of postcricoid hemangioma reprinted from Discolo and Koltai.1 A and B, Case 1. C and D, Case 2. E and F, Case 3.

tent. To test this premise, we analyzed a random selec- viewer’s ratings were different, the video was replayed until a tion of findings from office fiber-optic laryngoscopies of consensus was reached. infants and children to see if the presence of the cush- Guided by the consistent findings in the anatomical litera- ion could be consistently confirmed, how often it had any ture of a relatively larger plexus of vessels in fetuses and in- vascular coloration, and if it is more frequently seen in fants compared with those in older children and adults, we performed a logistic regression analysis on our data to determine younger compared with older children. We also per- the age at which the probability of seeing a cushion exceeds formed a search of the literature on the assumption that the probability of not seeing one. Based on this analysis, the there exists a body of knowledge in other subspecialty patients were then divided into 2 groups according to the de- fields that may provide a unified understanding of the rived age. Statistical analysis was performed using IBM SPSS postcricoid cushion. Statistics software (version 18; IBM Corp).

METHODS LITERATURE REVIEW

LARYNGOSCOPIC ANALYSIS A comprehensive literature review was performed, including anatomical, radiologic, and otolaryngologic publications and After institutional review board approval, we selected and re- texts. PubMed, Web of Science, and Google Scholar were que- viewed 125 consecutive flexible fiber-optic laryngoscopic (FFL) ried with the terms “postcricoid,” “postcricoid enlargement,” examinations on patients from birth to age 17 years were se- “postcricoid veins,” “postcricoid hemangioma/vascular mal- lected from our archived collection. FFL is routinely per- formation,” “cervical esophageal hemangioma,” and “pharyn- formed in our pediatric otolaryngology outpatient clinic as part geal plexus.” The references of all relevant articles were also of the workup for a variety of disorders of the pharynx and lar- obtained and reviewed, including historical texts and foreign- ynx. Videos are recorded and stored using KayPentax soft- language publications. ware and then stored on labeled CD-ROM discs (KayPentax). Videos were viewed in real time and in slow motion for the pres- RESULTS ence or absence of engorgement of the postcricoid region. If seen, we separately rated the fullness “small,” “medium,” or “large.” Any vascular-appearing violaceous discoloration of the FFL ANALYSIS postcricoid region was also looked for and was categorized as “none,” “transitional,” or “obvious.” Each video was judged in- A total of 125 videos of office-based FFL examinations dependently by each reviewer. On all videos for which the re- in pediatric patients were reviewed. Patient ages ranged

ARCH OTOLARYNGOL HEAD NECK SURG/ VOL 138 (NO. 6), JUNE 2012 WWW.ARCHOTO.COM 563

Presenting Symptoms No. (%) Stridor 38 (31.9) Hoarseness 36 (30.3) Snoring 22 (18.5) Noisy breathing 5 (4.2) Cough 5 (4.2) Weak voice 4 (3.4) Nasal obstruction 3 (2.5) Dysphagia 3 (2.5) Dyspnea 2 (1.7) Cyanotic spells 2 (1.7) Velopharyngeal insufficiency 1 (0.8) Sore throat 1 (0.8) Recurrent pneumonia 1 (0.8) Globus sensation 1 (0.8) Failure to thrive 1 (0.8)

Table 2. Final Diagnosis After Laryngoscopy in 125 Patients

Diagnoses No. (%) Laryngomalacia 23 (18.4) Normal 21 (17.6) Sleep apnea 17 (14.3) Vocal cord paralysis 16 (13.4) Gastroesophageal reflux disorder 11 (9.2) Vocal cord nodule 10 (8.4) Adenoid hypertrophy 8 (6.7) Laryngeal web 4 (3.4) Episodic croup 2 (1.7) Hemangioma 2 (1.7) Papillomatosis 2 (1.7) Subglottic stenosis 2 (1.7) Caustic ingestion 1 (0.8) Epidermolysis bullosa 1 (0.8) Laryngeal cleft, type II 1 (0.8) Posterior glottic stenosis 1 (0.8) Paradoxical vocal cord movement 1 (0.8) Tracheal stenosis 1 (0.8) Figure 2. Illustration of postcricoid venous plexus from Bourgery and Vocal cord cyst 1 (0.8) Jacob,12 reproduced with permission from Taschen GmbH.

from 3 weeks to 17 years, with an average age of 4.6 years and a median age of 2.3 years. The most common pre- tients with a medium (3.0%), small (3.3%), or absent (0%) senting symptom for laryngoscopy was stridor (Table 1), cushion (Figure 3). and the most common final diagnosis was laryngomala- Logistic regression demonstrated that the age at which cia (Table 2). There was no association of the present- the probability of seeing a cushion exceeds the probabil- ing symptoms or final diagnosis in clinic with the pres- ity of not seeing one is 24 months (2 years). Therefore, ence of a postcricoid cushion. patients were divided into 2 groups: those younger than Of the 125 videos, 119 had technically adequate vi- 2 years and those 2 years or older. There were 56 pa- sualization of the postcricoid region, while 6 were con- tients younger than 2 years, 49 (88%) of whom had a vis- sidered unusable owing to inability to assess the post- ible postcricoid cushion; 14% had a large cushion, 45% cricoid space. The postcricoid cushion was seen in 61.3% had a medium cushion, and 29% had a small cushion. of videos and rated as large in 8.4% of all patients, me- In this group, 13% did not have a visible cushion dium in 27.7%, and small in 25.2%. It was not present (Figure 4). There were 63 children in the group that in 38.7%. was 2 years or older; 24 (38%) had a visible postcricoid Most children (74.8%) did not have any vascular col- cushion, 3% had a large cushion, 13% had a medium cush- oration of the postcricoid region. An obvious bluish- ion, and 22% had a small cushion. In this group 62% did purple color was seen in 4.2% of patients, and 21.0% were not have a visible cushion. The infants younger than 2 rated as transitional. Those patients with a large postcri- years had a significantly higher rate of postcricoid cush- coid cushion were much more likely to have an obvious ions than children 2 years and older (PϽ.001; 2-tailed bluish-purple discoloration (30.0%), compared with pa- Fisher exact test).

ARCH OTOLARYNGOL HEAD NECK SURG/ VOL 138 (NO. 6), JUNE 2012 WWW.ARCHOTO.COM 564

100

Vascular/Blue Color of Postcricoid Cushions, % Vascular/Blue 10

0 Large Medium Small None Postcricoid Cushion Size

Figure 3. Vascular coloration of the postcricoid cushion based on size.

100 Age <2 y 90 Age ≥2 y 80 70 60 50

Patients, % 40 30 20 10 0 Any None Large Medium Small Postcricoid Cushion Figure 5. Illustration of postcricoid venous plexus, reproduced with assistance from Stanford Medical History Center (http://lane.stanford.edu Figure 4. Presence of a postcricoid cushion in relation to patient age. /med-history/index.html) from von Luschka.15

REVIEW OF LITERATURE the narrowest part of the pharynx. Unaware of a mus- cular upper esophageal sphincter, he suggested that the The vascular anatomy of the postcricoid region has been distension of these veins forms a barrier between the phar- well defined in the literature, however in a “frag- ynx and cervical esophagus. mented” fashion, with each specialty referring primar- In 1887, Bimar and Lapeyre16 described a remarkable ily to its own canon. A comprehensive review divided by venous plexus consisting of veins measuring 1 to 3 mm specialty follows. and connected by many anastomoses belonging to the lower or laryngeal pharynx. They noted that in contrast Anatomic Literature with other venous networks that progressively enlarge with age, at this site the reverse was true: the pharyn- The 1854 anatomy text of Bourgery and Jacob12 has al- geal venous plexus was proportionally larger and better ready been cited, but historically, it was Jean Cru- developed in infants than in older children and adults. veilhier13,14 (1791-1874), the French anatomist, who first In 1918, another German anatomist, Curt Elze,17 fur- described a very considerable pharyngeal venous plexus ther refined the detail of the postcricoid venous supply. in the postcricoid and posterior hypopharynx in 1834. He noted valves in the plexus that direct blood flow cra- In 1871, the German anatomist, Hubert von Luschka15 nially and described a “meandering” network receiving (1820-1875), provided a more detailed description of this tributaries from the mucosa and draining into the supe- plexus, including the first illustration, and termed it the rior laryngeal veins. He described this venous system as “pharyngolaryngeal plexus” (Figure 5). He noted that a “rete mirabile,” an anatomical term for a complex mesh- the plexus had a ventral portion on the dorsal aspect of work of similar vessels, and provided an illustration of the cricoid cartilage and a dorsal portion on the poste- the network. In a 1919 follow-up study, Elze and Beck18 rior pharyngeal wall. He observed that the plexus lies at demonstrated that the posterior cricoid rete mirabile is

ARCH OTOLARYNGOL HEAD NECK SURG/ VOL 138 (NO. 6), JUNE 2012 WWW.ARCHOTO.COM 565

©2012 American Medical Association. All rights reserved. Downloaded From: https://jamanetwork.com/ on 09/27/2021 present only on the posterior surface of the cricoid and cricoid venous plexus, which they compared with the not on the adjacent pyriform sinuses. They also de- pseudostrangulation sign seen at autopsy in patients with scribed a plexus on the posterior pharyngeal wall, which congestive heart failure. is more inferior than the posterior cricoid plexus, and There is an odd report by Friedland and Filly26 of a suggested that the engorgement of the opposing venous postcricoid impression on barium esophagram, which led networks formed a valve separating the pharynx and to the erroneous diagnosis of an esophageal tumor. Sur- esophagus. gical exploration revealed no mass, and findings from a In 1942, Baston,19 while studying the vertebral veins postoperative esophagram again demonstrated the im- via latex injections of the dorsal phallic vein of male ca- pression, which was then determined to be a normal davers, found that in several specimens the veins of the finding. pharynx were also filled. To study this further, he in- Dodds et al27 reviewed findings from normal radio- jected yellow oil paint, thinned with turpentine, into the graphic swallows and noted a posterior cricoid impres- veins of the inferior constrictor muscle. He found a rich sion, distinct from cervical esophageal webs. plexus of veins in the postcricoid region and posterior Allen et al28 reviewed fluoroscopic swallowing evalu- pharyngeal wall, confirming description by Elze17 and Elze ations in adults with and without dysphagia. They iden- and Beck18 of the postcricoid plexus being higher than tified a posterior cricoid “plication,” distinct from a web the plexus in the posterior pharyngeal wall. or cricoid arch impression, in a third of all patients, at- In 1951, Butler20 studied the pharyngeal venous tributing it to prolapsing mucosal folds over the muscles anatomy of human fetuses and adult autopsy specimens of the postcricoid region. They noted that the plication with injected India ink or a cast producing neoprene. He often changed shape during the swallow. provided a description of a pharyngolaryngeal venous Schmalfuss et al29 described the normal appearance plexus, with the ventral portion in the submucosa of the of the postcricoid region on computed tomographic (CT) dorsal surface of the cricoid cartilage, lying on the me- scans and magnetic resonance imaging (MRI) and noted dial part of the oblique and transverse arytenoid muscles marked enhancement of the mucosa in the postcricoid and the tendon of origin of the longitudinal muscle of region in 87% with CT imaging and 78% of patients on the esophagus. The plexus formed “two longitudinal MRI, with no significant differences based on age or sex. masses”20(p281) on each side of the cricoid midline sepa- rated by a gap of 2 to 6 mm. Cranially, the 2 halves of Otolaryngologic and Endoscopic Literature the plexus were “united by numerous cross anastomoses”20(p281) embedded in a mass of mucous glands. The In the otolaryngologic literature, descriptions of the vas- veins were confined to the postcricoid region and did not cularity of the postcricoid region are mostly confined to extend into the pyriform sinuses. Interestingly, Butler20 case reports and small case series of postcricoid heman- referenced an otolaryngologist, A. S. H. Walford, who de- gioma. A retrospective review of the published cases, in- scribed the bluish “bolster-like” swelling in the hypo- cluding the clinical photographs, reveals that in some pharynx of a living person during mirror laryngoscopy. cases, the findings are more comparable with an en- From 1984 to 1985, using latex and then India ink gorged postcricoid cushion than with a hemangioma, and injections, Tose et al23 visualized the anterior and pos- often behaved as such (Table 3). terior pharyngeal venous plexus and further delineated The first clinical report of a postcricoid hemangioma the area into venous networks and “pads.” They specu- was by Goldsmith et al,2(p851) who described a “dark, vas- lated that the pads may become turgid during inspira- cular appearing mass in the postcricoid area,” which “dra- tion and thereby obstruct the entrance of air into the matically increased in size” with crying, and caused com- esophagus. plete obstruction of the esophageal inlet leading to failure More recently, Ramaekers et al,24 in a 1990 anatomi- to thrive. After tracheotomy, the lesion was excised using cal description of the pharyngoesophageal transition zone, carbon dioxide laser; the pathologic findings were con- suggested that the orientation of the venous plexus pro- sistent with a capillary hemangioma. vides rigidity during swallowing for easier passage of the Tonsakulrungruang3 reported a dark blue mass on the bolus into the esophagus. They also noted a striking dif- posterior surface of the cricoid, just above the cricopha- ference between the plexus in the fetus and adult, with ryngeal opening. Angiography showed that the mass was the fetal plexus being an extensive venous labyrinth with supplied by the left superior thyroid artery; the patho- numerous fenestrae and in the adult a thin system of lon- logic findings were consistent with hemangioma. gitudinal parallel veins. In a letter to the editor in 1997, Gyokeres et al4 described a hemangioma on the retrocricodeal area that ex- Radiographic Literature tended to the anterior cervical esophagus, appearing during retching, moving in peduncular fashion. Radiologists have consistently noted a convexity of tis- Ludemann and Kozak5 described 4 infants who pre- sue behind the cricoid on swallow studies. Pitman and sented with stridor and dysphagia and who were noted Fraser,25 in a study of barium swallow examinations for to have swelling of the esophageal inlet mucosa, occa- dysphagia, identified a “postcricoid impression” in 86% sionally with dark reddish-purple staining, that was most of patients with dysphagia and 90% of control patients. prominent while the infants were crying. In at least 1 case, Considering this to be a normal finding, they suggested the lesion was difficult to visualize in the operating room. that it is a prolapse of mucosa over a venous plexus. Ad- Desuter et al6 reported the cases of 2 infants with post- ditional injection studies demonstrated a distended postcricoid hemangioma presenting with dysphagia. One had

ARCH OTOLARYNGOL HEAD NECK SURG/ VOL 138 (NO. 6), JUNE 2012 WWW.ARCHOTO.COM 566

©2012 American Medical Association. All rights reserved. Downloaded From: https://jamanetwork.com/ on 09/27/2021 Table 3. Summary of 26 Reported Cases of Postcricoid Hemangioma and Vascular Malformations in the Literaturea

Other Cyclical Presented With Vascular Steroid Treatment Difficulty Enlargement Patients, Patient Anomalies Visualizing With Source No. Agesb Dysphagia Stridor/CyanosisNoted Systemic Intralesional in OR Crying/Straining Goldsmith et al,2 1 16 mo 1 NA NA NA NA NA Yes 1987 Tonsakulrungruang,3 134y 1 NA NANANA NA NA 1988 Gyokeres et al,4 1 18 y NA NA NA NA NA NA NA 1997 Ludemann and 4 2d,2d,7wk, 3 4 NA NA NA Yes Yes Kozak,5 2003 3mo Desuter et al,6 2004 2 10 d, 24 mo 2 2 NA 2 NA NA NA Discolo and Koltai,1 4 8 mo, 8 mo, 2 3 NA NA NA NA NA 2004 10 mo, 15 mo Zur et al,7 2005 4 8 mo, 1 y, 3 y, 321NANAYesYes 7y Awwad and 4 3 mo, 9 mo, NA 3 NA 1 NA Yes Yes Mortelliti,9 2006 16 mo, 6 y Folia et al,10 2007 4 6 wk, 2 mo, 2 4 NA 3 1 NA NA 2mo,4mo Sternbach et al,11 1 7 mo 1 NA NA 1 NA NA Yes 2010 Total, No. (%) 26 Range, 15 (57) 18 (69) 1 (4) 7 (27) 1 (4) NA NA 2dto34y

Abbreviations: NA, not applicable; OR, operating room. aSeveral reports note cyclical enlargement with crying and straining, and difficulty visualizing in the OR, although not all patients underwent operative endoscopy. bMedian age, 8 months.

an MRI scan, which showed a hypervascular lesion at the Awwad and Mortelliti9 reported the cases of 4 patients approximate level of the dorsal pharyngeal plexus. Both who on flexible laryngoscopy had a bluish expansile le- patients underwent systemic steroids with methylpred- sion in the postcricoid region, with the “propensity to en- nisolone, with subsequent regression of the lesion and large with crying or straining.”9(p191) They described a pa- resolution of the dysphagia. tient with a significant postcricoid lesion while crying, Discolo and Koltai1 reported the cases of 4 children, although “when the child was calmed with a pacifier, the ages 8 to 15 months, with postcricoid hemangioma. These mucosal bulge diminished substantially.”9(p192) Three pa- patients were referred to the otolaryngology depart- tients had operative airway evaluation, during which the ment for different reasons, including stridor and aspira- mass was often difficult to identify. In 1 case, the lesion tion, dysphagia, and laryngomalacia. The initial diagno- was difficult to visualize in the operating room, but a flex- sis was made on office flexible laryngoscopy, with which ible laryngoscopy was performed in the recovery room with the vascular lesions in the postcricoid region were seen. the patient crying again demonstrated a large, expansile All patients underwent operative airway evaluation un- hemangioma. The MRI scans in 1 patient did not reveal der anesthesia, which confirmed the presence of the le- any vascular abnormalities. One patient was followed for sions. All 4 children were treated conservatively with ob- 6 years with serial laryngoscopy, with a substantial regres- servation, with no further sequelae. sion in the size of the lesion. Zur et al7(p1698) reported 4 cases of postcricoid heman- Folia et al10 described 4 cases of postcricoid hemangioma, including a 3-year-old child with dysphagia who, gioma, 3 of which were treated with open excision after while crying, had “ballooning of the postcricoid mass into failing treatment systemic steroids. In 1 case, an MRI scan a blush, grape-like vascular appearing mass.” A swallow showed an encircling lesion in the cervical esophagus on study confirmed the ballooning effect during crying. Find- T2 and T1 with gadolinium. Pathology results from the ings from an MRI scan were unrevealing, and Valsalva excisional cases are not provided. and Trendelenburg positioning was necessary to dem- Sternbach et al11 reported a 7-month-old with cough- onstrate the mass during operative endoscopy. The other ing spells while feeding. She had a round bluish mass in 3 patients also had ballooning with Valsalva positioning the postcricoid area during crying or straining, but it dis- or cry, and 3 of the 4 underwent excision with potassium- appeared when she relaxed. She was treated with sys- titanyl-phosphate laser or carbon dioxide laser. The patho- temic steroids for 1 month, and the lesion was no longer logic findings of the resected tissue on 1 of their pa- seen at 18 months. tients with extensive venous malformation (VM) and Finally, in an recent study by Parhizkar et al,30 pa- lymphatic malformations were consistent with a VM with tients with vascular anomalies of the airway were com- an increased number of small stromal vessels and mul- pared. Specifically, the characteristics of airway infan- tiple, dilated vascular channels. tile hemangiomas (IHs), a vascular tumor, are contrasted

ARCH OTOLARYNGOL HEAD NECK SURG/ VOL 138 (NO. 6), JUNE 2012 WWW.ARCHOTO.COM 567

Efferent vein Site of mucous gland Pyriform sinus Ventral pharyngolaryngeal venous plexus Afferent vein

Figure 7. Illustration of bilateral postcricoid venous plexus, reproduced from Bassett and Gruber,21 with permission from Stanford Medical History Center (http://lane.stanford.edu/med-history/index.html). Figure 6. Demonstration of bilateral postcricoid venous plexus after neoprene injection, reproduced from Butler.20 Reproduced with permission from BMJ Publishing Group Ltd. In the videos, the recurring dilation of the postcricoid cushion, which was most prominent during the ex- with those of airway VMs, a vascular malformation. They piratory phase of crying, is well visualized in nearly ev- found 11 patients with IHs and 6 with VMs. Those with ery infant. The cycle of the cry begins with a large IHs had a younger age at presentation, all presented with inspiratory effort, with hypopharyngeal and supraglot- stridor, and typically had associated cutaneous IHs. The tic dilation, a descent of the larynx, and maximal abduc- IHs were in the glottis or subglottis. The VMs were lo- tion of the vocal cords. The postcricoid cushion at this cated in the postcricoid (in 6 of 6 patients) and epiglot- phase can generally be visualized as a rounded mound tic regions, and the patients presented at an older age and covered with loose mucosa. As the expiratory phase of were less likely to require intervention. The distinction the cry begins, the vocal cords adduct, and the larynx be- is confirmed with erythrocyte-type glucose transporter gins to rise as the hypopharynx constricts around it. As protein (GLUT-1) testing, which is a useful immunohis- the cry emerges, the supraglottic folds constrict, and the tochemical marker for IHs. It has been shown that IHs postcricoid region becomes full and prominent. Expand- have intense endothelial GLUT-1 immunoreactivity during tightly against the posterior and posterolateral hy- ing all phases of these lesions (proliferation, plateau, in- popharyngeal wall, which symmetrically appears to ac- volution), but other vascular malformations, including tively tighten toward it, the cushion presses both VMs, have no GLUT-1 immunoreactivity.31 In the study posteriorly as well as laterally at a level overlying the pos- by Parhizkar et al,30 the IHs stained positive for GLUT-1, terior lamina of the cricoid cartilage. The carotid arter- and the VMs were negative for GLUT-1, as expected. ies, which pulsate in the hypopharynx just lateral to the cushion, medialize during the cry and are tightly op- COMMENT posed by the cushion (Video 2). In most circumstances the shape of the cushion is ovoid, but at times it appears to have a midline raphe and is bilobed (Video 3). This is Buttressed by our study and review of the literature, the consistent with the dissections of Butler20 and Bassett and analysis and interpretation of our findings are based on Gruber,21 which show the separate but connected bilat- the following propositions: eral plexus of veins at this site (Figure 6 and Figure 7). 1. There is a cyclical engorgement of the postcricoid A modest 4.2% of the videos demonstrate a bluish to vio- region, most notably during the expiratory phase of an laceous discoloration of the cushion that is more notice- infant’s cry. We term this phenomenon “the postcricoid able during expiration (Video 4). With extreme agita- cushion.” tion, the entire hypopharynx constricts around the larynx, 2. The postcricoid cushion is more prominent in infants and the posterior pharyngeal wall overrides the cushion and less noticeable or not observed in older children. obliterating its view. 3. Vascular coloration of the postcricoid cushion is The postcricoid cushion can also be demonstrated on infrequently but regularly observed. direct microlaryngoscopy by lifting the larynx forward 4. There is a well described venous plexus that cor- with a horizontally held right angle probe (Figure 8 and responds to the site of the postcricoid cushion. Video 5). What becomes apparent is the posterior as- 5. This plexus diminishes in relative size with initial pect of the circumferential vascular plexus mentioned by growth and maturation. von Luschka, Elze, and others.15-24 The posterior com- 6. Vascular anomalies of the postcricoid region are ponent is less prominent than the cushion and some- rare. what caudal to it. This region is harder to see on our fiber- 7. Stridor and dysphagia are the most common symp- optic examinations because the cushion, being more toms of postcricoid vascular anomalies. superior, overlaps it during engorgement.

ARCH OTOLARYNGOL HEAD NECK SURG/ VOL 138 (NO. 6), JUNE 2012 WWW.ARCHOTO.COM 568

©2012 American Medical Association. All rights reserved. Downloaded From: https://jamanetwork.com/ on 09/27/2021 Figure 8. Cushion with vascular discoloration seen during operative Figure 9. Vascular anomaly of postcricoid region expanding during microlaryngoscopy. microlaryngoscopy.

We propose that the plexus of veins overlying the pos- hemangioma, and Tonsakulrungruang3 described patho- terior cricoid cartilage, so well documented in the ana- logic findings consistent with hemangioma. The prob- tomical literature, is the same mucosally covered entity lem is that in the late 1980s there was much uncertainty that forms the postcricoid cushion we observe during flex- regarding the classification of vascular malformations and ible laryngoscopy in our patients and that is seen radio- what we classify today as VMs were often referred to as graphically during swallowing. The locations corre- capillary hemangiomas. Several reports describe involu- spond. The variability in morphologic characteristics, with tion with systemic steroid treatment, but the same re- prominence of the cushion’s midline raphe, is in agree- ports also site loss of visualization with maturation.6,9-11 ment with the dissections. The intermittent violaceous Zur et al7 describe the treatment dilemma of infants discoloration infers a vascular relationship. The promi- seen for dysphagia who had postcricoid hemangiomas nence of the cushion in our youngest patients parallels or vascular malformations diagnosed on office fiber- the relative robustness of the plexus in fetal dissections. optic laryngoscopy, yet whose lesions were difficult to There is also a dynamic physiologic correspondence be- visualize in the operating room unless Valsalva and Tren- tween the cushion and the plexus. delenburg measures were performed. They questioned During the expiratory phase of an infant’s cry, there why a postcricoid hemangioma is pressure dependent, is a sharp increase in intrathoracic pressure with a con- whereas a subglottic hemangioma is not, and suspected current decrease in venous return into the right atrium. that these lesions represent vascular malformations in- The resulting venous congestion of the head and neck, stead. It has been our observation that VMs, in general, most recognizable in the plethoric facial discoloration of become more prominent during anesthesia. This sug- a wailing infant, is coincidental with the ballooning promi- gests that malformations of the postcricoid region may nence of the cushion during the cry. Indeed, the case re- be expected to behave this way. In 1 child of our origi- ports repeatedly described a cyclical engorgement and nal 4 the lesion became more distended under anesthe- discoloration associated with the patient cry, which re- sia compared with what was seen on office fiber-optic en- solved on relaxation.2,5,7,9,11 doscopy1 (Figure 9). In retrospect, today we would call The correspondence of location between the postcri- this a postcricoid VM and not a hemangioma. We would coid plexus, cushion, and published descriptions of post- also suggest that several of the previous case reports of cricoid vascular anomalies strongly implies the plexus postcricoid hemangioma may actually have been post- as the source of those lesions. However, the review of the cricoid VMs, or a large postcricoid cushion with vascu- literature on postcricoid hemangiomas suggests 2 sepa- lar coloration. rate entities: The pathologic uncertainty has begun to be resolved by Parhizkar et al,30 using the IH-specific immunohisto- 1. A vascular stained postcricoid cushion, visually en- chemical marker GLUT-1. They confirm that there are gaging but asymptomatic. distinct differences in the examination and presentation 2. A true vascular anomaly causing dysphagia and/or of hemangiomas and VMs, and this is borne out with airway obstruction. GLUT-1 staining. Similarly, propranolol will also help Our study suggests that the former are a fairly com- resolve this question.32 Until then, we propose the term mon physiologic phenomenon, whereas the latter are both “postcricoid vascular anomaly” (PCVA) to bridge the un- rare and pathologic. The question that then remains is certainty. whether the true vascular anomalies are indeed heman- Ludemann and Kozak8 are correct when they suggest giomas or are instead VMs. While it is reasonable to as- that the red or purple mass seen at the esophageal inlet sume that a hemangioma can occur at this site, the patho- which engorges during cry is likely due to increased in- logic evidence is not convincing. Goldsmith et al2 trathoracic pressure, and that postcricoid prolapse is not described pathologic findings consistent with capillary a surgical disease. However, in the literature there are sev-

ARCH OTOLARYNGOL HEAD NECK SURG/ VOL 138 (NO. 6), JUNE 2012 WWW.ARCHOTO.COM 569

©2012 American Medical Association. All rights reserved. Downloaded From: https://jamanetwork.com/ on 09/27/2021 eral case reports of infants who presented primarily with note that there were no episodes of gastric regurgitation dysphagia and have an identifiable, usually large, post- on any of our videos. cricoid vascular lesion, not what we are describing as a The weakness of our study is that it is neither pro- normal postcricoid cushion.2,5-7,11 The cause of dyspha- spective nor randomized. The observers who viewed and gia has to be inferred, but a mechanism based on an ex- scored the videos were not blinded or disinterested in the cess of vascular tissue above the esophageal inlet caus- topic. We acknowledge these limitations and look for- ing a “ball-valve” type obstruction seems plausible. ward to our results being tested with further studies. Evidence for this effect is found in the radiologic literature, where there are consistent descriptions of a ridge CONCLUSIONS of tissue, distinct from the arch of the cricoid cartilage, which forms an impression or plication in the postcricoid region on esophagram.25-29 This is considered a nor- We have analyzed the fiber-optic laryngoscopy videos mal anatomic finding corresponding to the postcricoid of 119 children independently selected from our clinic venous plexus seen on adult anatomical dissections and archive and found a consistent cyclical engorgement of presumably to the postcricoid cushion as well. The nor- the posterior cricoid region, occasionally stained viola- mal plication is considered to be dynamically passive, ceous, that is most prominent during crying and most being drawn into the esophageal inlet during the nega- conspicuous in the first 2 years of life. This is the clini- tive pressure phase of the hypopharyngeal swallow. The cal counterpart of the well described venous plexus cause of dysphagia with PCVA is a consequence of this overlying the postcricoid region and is relatively larger same effect; however, it is magnified by their greater scale in the infant. We term this structure the postcricoid compared with normal plications, resulting in a ball- cushion. valve barrier to passage of the bolus. The cushion is clinically related to, yet distinct from, The ball-valve mechanism suggests that reducing the the true vascular anomalies that occur at this site. A bet- size of the PCVA should diminish symptoms. Based on ter appreciation of the normal physiologic phenom- the case reports in the literature, time seems to be effec- enon of crying and the anatomic basis for postcricoid tive in most cases. A few authors have reported success pathologic characteristics may provide direction for ef- with systemic steroids and surgical excision. Because fective therapy when symptoms warrant. We propose we have had no experience actively treating PCVAs we that the function of the cushion is to augment the upper have no recommendations in this regard. Attempt at and lower esophageal sphincter in preventing emesis medical management seems reasonable, but without a and gastric regurgitation from elevated intra-abdominal histologic diagnosis, knowing what to treat becomes pressure during the expiratory phase of crying. Our in- problematic. terpretation weaves together multiple threads of evi- Three theories about the function of the posterior cri- dence from a variety of sources into a coherent and uni- coid venous plexus have been previously proposed: fying narrative of the vascular anatomy, pathologic characteristics, and physiologic mechanisms, of the 1. Protection from aspiration. The cushion facili- postcricoid region. tates the bifurcation of the bolus during swallowing to flow lateral to the larynx, particularly at the level of the interarytenoid notch, with the 2 streams then coming Submitted for Publication: June 2, 2011; final revision back together at a level inferior to the larynx. This received September 21, 2011; accepted April 17, 2012. would be particularly important in the neonate and ex- Correspondence: Stephen R. Hoff, MD, Division of Pe- plains why the pad is proportionally larger in the diatric Otolaryngology, Children’s Memorial Hospital, younger population.17,20 2300 Children’s Plaza, PO Box 25, Chicago, IL 60614 2. Protection from regurgitation. The venous plexus ([email protected]). adds a cephalic myoangiogenic component to the upper Author Contributions: Drs Hoff and Koltai had full ac- esophageal sphincter.17 cess to all the data in the study and take responsibility 3. Prevention of aerophagia during crying.23 for the integrity of the data and the accuracy of the data All of these theories derive from anatomical dissec- analysis. Study concept and design: Koltai. Acquisition of tions. Yet from our clinical perspective, the physiologic data: Hoff and Koltai. Analysis and interpretation of data: phenomenon of the postcricoid cushion is most dynami- Hoff and Koltai. Drafting of the manuscript: Hoff and Kol- cally expressed during an infant’s cry. With the expira- tai. Critical revision of the manuscript for important intel- tory phase of the cry, nearly every muscle of the child is lectual content: Koltai. Statistical analysis: Hoff. Admin- tensed, and there is a sharp increase in both intra- istrative, technical, and material support: Koltai. Study abdominal and intragastric pressure. The upper and lower supervision: Koltai. esophageal sphincters restrict gastric content within the Financial Disclosure: None reported. stomach during these dynamic pressure changes.33 On Previous Presentation: This study was presented at the endoscopy, there is a tight compression of the cushion 2011 American Society of Pediatric Otolaryngology against the posterior hypopharynx at the peak of the cry. (ASPO) Annual Meeting; April 30, 2011; Chicago, Illi- We propose that the postcricoid cushion adds a third layer nois; and received third place for the Charles F. Fergu- of more proximal protection against emesis during the son Clinical Research Award. cyclical periods of maximal intra-abdominal pressure that Online-Only Material: The videos are available at http: occur in the expiratory phase of a cry. It is interesting to //www.archoto.com.

ARCH OTOLARYNGOL HEAD NECK SURG/ VOL 138 (NO. 6), JUNE 2012 WWW.ARCHOTO.COM 570

©2012 American Medical Association. All rights reserved. Downloaded From: https://jamanetwork.com/ on 09/27/2021 16. Bimar L, Lapeyre JM. Recherches sur les veines du pharynx. Comp Rend Acad REFERENCES dSc. 1887;105:825. 17. Elze C. Die venosen Wundernetze der Pars laryngea pharyngis. Anat Anz. 1918; 1. Discolo CM, Koltai PJ. Postcricoid hemangioma presenting as dysphagia: a re- 51:205-207. port of 4 cases. Arch Otolaryngol Head Neck Surg. 2004;130(12):1420-1422. 18. Elze C, Beck K. Die venosen Wundernetze der Hypopharynx. Z Ohrenheilkunde. 2. Goldsmith MM III, Strope GL, Postma DS. Presentation and management of post- 1919;77:185-197. cricoid hemangiomata in infancy. Laryngoscope. 1987;97(7, pt 1):851-853. 19. Baston O. Veins of the pharynx. Arch Otolaryngol. 1942;36(2):212-219. 3. Tonsakulrungruang K. Postcricoid hemangioma. J Med Assoc Thai. 1988;71(8): 20. Butler H. The veins of the oesophagus. Thorax. 1951;6(3):276-296. 466-468. 21. Bassett D, Gruber W. Stereoscopic Images of Human Anatomy. Philadelphia, PA: 4. Gyokeres T, Lippai G, Dekany K. Haemangioma in the hypopharynx: necessity Sawyers; Williams and Wilkins; 1962:plate 83-3. to screen the larynx and hypopharynx during upper GI endoscopy. Endoscopy. 22. Tose D, Rodrigues H, DiDio LJ. The venous architecture of the human pharyngo- 1997;29(9):887. esophageal transition. Arch Ital Anat Embriol. 1984;89(3-4):157-165. 5. Ludemann JP, Kozak FK. Postcricoid prolapse: an unusual sign of paediatric air- 23. Tose D, Rodrigues H, DiDio LJ. Mucosal and submucosal veins of the human way disease. J Otolaryngol. 2003;32(2):128-131. pharyngo-esophageal transition. Arch Ital Anat Embriol. 1985;90(1):9-15. 6. Desuter GR, El Makhloufi K, François GJ, et al. Postcricoid hemangioma: an over- 24. Ramaekers D, Mebis J, Geboes K, Desmet V. Vascularization of the pharyngo- looked cause of dysphagia in infants? a case report. Dysphagia. 2004;19(1): esophageal transition zone. Acta Gastroenterol Belg. 1990;53(4):376-385. 48-51. 25. Pitman RG, Fraser GM. The postcricoid impression on the esophagus. Clin Radiol. 7. Zur KB, Wood RE, Elluru RG. Pediatric postcricoid vascular malformation: a di- 1965;16:34-39. agnostic and treatment challenge. Int J Pediatr Otorhinolaryngol. 2005;69(12): 26. Friedland GW, Filly R. The postcricoid impression masquerading as an esoph- 1697-1701. ageal tumor. AmJDigDis. 1975;20(3):287-291. 8. Ludemann JP, Kozak FK. Letter to the editor regarding pediatric postcricoid masses. 27. Dodds W, Stewart E, Logemann J. Physiology and radiology of the normal oral Int J Pediatr Otorhinolaryngol. 2006;70(8):1499-1501. and pharyngeal phases of swallowing. Am J Roentgenol. 1990;154(5):953-963. 9. Awwad RJ, Mortelliti AJ. Postcricoid hemangioma of childhood: report of four 28. Allen J, White C, Leonard R, Belfasky P. Posterior cricoid region fluoroscopic cases. Ann Otol Rhinol Laryngol. 2006;115(3):191-194. findings: the posterior cricoid plication. Dysphagia. 2011;26(3):272-276. 10. Folia M, Naiman N, Dubois R, Ayari S, Froehlich P. Management of postcricoid 29. Schmalfuss IM, Mancuso AA, Tart RP. Postcricoid region and cervical esophagus: and upper esophageal hemangioma. Int J Pediatr Otorhinolaryngol. 2007;71 normal appearance at CT and MR imaging. Radiology. 2000;214(1):237-246. (1):147-151. 30. Parhizkar N, Manning SC, Inglis AF Jr, Finn LS, Chen EY, Perkins JA. How air- 11. Sternbach D, Fishman G, Ben-Sira L, DeRowe A. Postcricoid vascular lesion in way venous malformations differ from airway infantile hemangiomas. Arch Oto- an infant. Harefuah. 2010;149(2):77-79, 125. laryngol Head Neck Surg. 2011;137(4):352-357. 12. Bourgery J, Jacob NH. Atlas of Human Anatomy and Surgery: The Complete Col- 31. North PE, Waner M, Mizeracki A, Mihm MC Jr. GLUT1: a newly discovered im- ored Plates of 1831-1854. 25th ed. Koln, Germany: Taschen; 2005:177. munohistochemical marker for juvenile hemangiomas. Hum Pathol. 2000;31 13. Cruveilhier J. Anatomie Descriptive. Vol 3. Paris, France: Bechet Jeune; 1834: (1):11-22. 286. 32. Leáute´-Labrèze C, Dumas de la Roque E, Hubiche T, Boralevi F, Thambo JB, Ta- 14. Cruveilhier J. The Anatomy of the Human Body. New York, NY: Harper & Bros; ëbı A. Propranolol for severe hemangiomas of infancy. N Engl J Med. 2008; 1844:349. 358(24):2649-2651. 15. von Luschka H. Der Kehlkopf des Menschen. Tubingen, Germany: H. Laupp; 1871: 33. Hirshberg J, Szende T, Koltai P, Illenyi A. Pediatric Airway: Cry Stridor and Cough. 147, Plate 8, Figures 3 and 4. San Diego, CA: Pleural Publishing; 2008.

ARCH OTOLARYNGOL HEAD NECK SURG/ VOL 138 (NO. 6), JUNE 2012 WWW.ARCHOTO.COM 571