The LaryngoscopeVC 2012 The American Laryngological,Rhinological and Otological Society, Inc.

First Bite Syndrome: Incidence, Risk Factors,Treatment, and Outcomes

Gary Linkov, BS; Luc G. T. Morris, MD; Jatin P. Shah, MD; Dennis H. Kraus, MD

Objectives/Hypothesis: First bite syndrome (FBS) refers to facial pain characterized by a severe cramping or spasm inthe parotid region with the first bite of each meal that diminishes over the next several bites.1,2 It is a potential sequela ofsurgery involving the infratemporal fossa (ITF), parapharyngeal space (PPS), and/or deep lobe of the parotid gland. The inci-dence, risk factors, treatment options, and outcomes of FBS are poorly understood. We hypothesized that certain clinical andtumor variables independently predict the development of FBS.

Study Design: Retrospective cohort study.Methods: We reviewed the records of 499 patients (mean age, 50 years; range, 12–81 years) undergoing surgery of the

deep lobe of the parotid gland, PPS, and/or ITF between 1992 and 2010. Minimum follow-up time was 3 months (median,39 months). Patient, tumor, and FBS characteristics were analyzed. Incidence was calculated using the Kaplan–Meier method.Univariate analyses and multivariate logistic regression were used to identify independent risk factors for FBS. Patients devel-oping FBS were interviewed to assess the efficacy of various treatment modalities.

Results: FBS developed in 45 patients (incidence, 9.6%), at a mean time of 97 (range, 6–877) days from surgery. Onmultivariate analysis, three variables were significant independent risk factors for FBS: sympathetic chain sacrifice (odds ratio[OR], 4.7; P ! .008), PPS dissection (OR, 8.7; P ! .001), and resection of only the deep lobe of the parotid gland (OR, 4.2; P !.002). FBS developed in 48.6% of patients undergoing sympathetic chain sacrifice, 22.4% of patients undergoing PPS dissec-tion, 38.4% of patients undergoing isolated deep lobe parotid resection, and 0.8% of patients undergoing total parotidectomy.Partial resolution of FBS symptoms occurred in 69% and complete resolution in 12%. Of 45 FBS patients, 15 (33%) under-went at least one type of treatment for symptomatic relief. No treatment consistently provided effective symptomatic relief.

Conclusions: The strongest independent risk factors for FBS are PPS dissection, deep lobe of parotid resection, andsympathetic chain sacrifice. Patients undergoing surgery with dissection and/or manipulation in these anatomical sites andstructures should be thoroughly counseled about the risk of developing FBS.

Key Words: First bite syndrome, parapharyngeal space, infratemporal fossa, pain, parotid.Level of Evidence: 2c

Laryngoscope, 122:1773–1778, 2012

INTRODUCTIONFirst bite syndrome (FBS) refers to facial pain char-

acterized by a severe cramping or spasm in the parotidregion with the first bite of each meal that diminishesover the next several bites. It is a potential sequela ofsurgery involving the infratemporal fossa (ITF), para-pharyngeal space (PPS), and/or deep lobe of the parotidgland. The constellation of symptoms varies from tran-sient and mild to persistent and severe.1–4 The anatomiclocation most frequently associated with the develop-ment of postoperative FBS is the parapharyngeal space,and the mechanism is believed to be a loss of sympa-

thetic innervation to the parotid gland.1,5 The incidenceof FBS after head and neck surgery remains unknown.

In addition, although the resection of certain struc-tures, such as the sympathetic chain, during surgeryhas been speculated to be linked with FBS,1,4,5 the fac-tors associated with FBS remain poorly characterized.We hypothesized that certain clinical and tumor varia-bles would independently predict the development ofFBS. Our objective was to define the incidence, potentialrisk factors, treatment, and outcomes of FBS followinghead and neck surgery.

MATERIALS AND METHODSThe records of 499 patients undergoing surgery in any of

three anatomic locations between 1992 and 2010 at MemorialSloan-Kettering Cancer Center (MSKCC) were analyzed. Fol-lowing institutional review board approval, we conducted acomprehensive search of operative reports and pathologicreports to identify candidate cases of surgery involving the deeplobe of the parotid gland, PPS, and ITF. These anatomic siteswere chosen to capture patients at risk of developing FBS basedon the current literature and mechanistic understanding of thiscondition. Patients with exposure to radiation of the head orneck before the operation, for any reason, were included. Bothbenign and malignant tumors were included. Exclusion criteria

From the Head and Neck Service, Department of Surgery, MemorialSloan-Kettering Cancer Center, New York, New York, U.S.A.

Editor’s Note: This Manuscript was accepted for publication March28, 2012.

An oral presentation of this article was given at the 2012 meetingof the North American Skull Base Society, Las Vegas, Nevada, U.S.A.,February 18, 2012.

The authors have no funding, financial relationships, or conflictsof interest to disclose.

Send correspondence to Luc G. T. Morris, MD, MSc, 1275 YorkAve #S-1210A, New York, NY, 10065. E-mail: morrisl@mskcc.org

DOI: 10.1002/lary.23372

Laryngoscope 122: August 2012 Linkov et al.: First Bite Syndrome

1773

included previous surgery in these anatomic sites prior to refer-ral to MSKCC, procedures limited to biopsy, and postoperativefollow-up of <3 months.

Patient factors, presenting symptoms, aspects of medicalhistory, details of the operative procedure, pathology, and

details of radiotherapy were recorded. Operative reports werereviewed with a particular focus on dissection in the sites of in-terest and ligation and/or resection of pertinent vascular and/ornervous structures. The integrity of the sympathetic chain,external carotid artery, and its branches was recorded. Postop-erative findings such as facial nerve palsy, Horner syndrome,temporomandibular joint pain, and FBS were recorded. Inpatients developing FBS, time of onset, treatment, and outcomeat last follow-up visit were recorded. Patients who developedFBS, and for whom current contact information was available,were individually interviewed by telephone with a structuredquestionnaire.

TABLE 1.Patient Characteristics and FBS Incidence.

Characteristic No. (%) Incidence (%) P

Total 499 (100) 45 (9)

Sex

Male 261(52) 14 (5) 0.003

Female 238 (48) 31 (13)

Prior RT

Yes 47(9) 0 0.023

No 452 (91) 45 (10)

Age, y

Mean 56 (2-91)

<45 124 (25) 15 (12) 0.112

45-58 125 (25) 15 (12)

59-69 125 (25) 9 (7)

70-91 125 (25) 6 (5)

Diabetes

Yes 40 (8) 1 (3) 0.133

No 459 (92) 44 (10)

Neuropathy

Yes 28 (6) 5 (18) 0.093

No 471 (94) 40 (8)

PVD

Yes 4 (1) 0 0.527

No 495 (99) 45 (9)

Psychiatric History

Yes 64 (13) 11 (17) 0.015

No 435 (87) 34 (8)

Anxiety

Yes 30 (47) 7 (23) 0.221

No 34 (53) 4 (12)

FN Fully Intact Preop

Yes 454 (91) 44 (10) 0.095

No 45 (9) 1 (2)

FN Grade Preop

Mean 1.2

1-3 478 (96) 45 (9) 0.140

4-6 21 (4) 0

Horner’s Syndrome Preop

Yes 4 (1) 0 0.527

No 495 (99) 45 (9)

FBS Preop

Yes 0 0 N/A

No 499 (100) 45 (9)

TMJ Pain Preop

Yes 12 (2) 1 (8) 0.933

No 487 (98) 44 (9)

Abbreviations: FBS, first bite syndrome; RT, radiation; PVD, periph-eral vascular disease; FN, facial nerve; TMJ, temporomandibular joint; N/A,not available.

TABLE 2.Tumor Characteristics and FBS Incidence.

Characteristic No. (%) Incidence (%) p

Side

Right 233 (47) 19 (8) 0.529

Left 266 (53) 26 (10)

Site of Origin

PPS 73 (15) 20 (27) <0.001

ITF 14 (3) 0

Parotid 257 (52) 25 (10)

Portion of Parotid

Deep 122 (47) 23 (19) <0.001

Superficial 7 (3) 0

Both 128 (50) 2 (2)

Intraparotid Node Status

Positive 28 (11) 0 0.066

Negative 229 (89) 25 (11)

Other 155 (31) 0

Site of interest Path " 114 (74) 0 N/A

Site of interest Path # 41 (26) 0

Histology

Schwannoma 28 (6) 10 (36) <0.001

Paraganglioma 41 (8) 9 (22)

Pleomorphic Adenoma 100 (20) 20 (20)

Other 330 (66) 6 (2)

Presence of Tumor

Yes 482 (97) 45 (9) 0.187

No 17 (3) 0

Tumor Type

Benign 217 (44) 43 (20) <0.001

Malignant 265 (53) 2 (1)

Size, cm

Mean 3.5 (0.5–13.4)

$2 89 (18) 5 (6) 0.331

2.1–3.1 90 (18) 6 (7)

3.2–4.5 90 (18) 10 (11)

4.5–3.4 91 (18) 11 (12)

Margins

Positive 139 (28) 12 (9) 0.760

Negative 145 (29) 12 (8)

Close (<5mm) 69 (14) 4 (6)

Abbreviations: FBS, first bite syndrome; PPS, parapharyngeal space;ITF, infratemporal fossa; Path, pathology; N/A, not available.

Laryngoscope 122: August 2012 Linkov et al.: First Bite Syndrome

1774

The cumulative incidence of FBS was calculated using theKaplan–Meier method, censoring patients at the time of last fol-low-up. Univariate analysis of patient, treatment, and tumorcovariates and FBS incidence was carried out using the chi-squared statistic and the log-rank test, with significance defineda priori as P < .05. Significant covariates were then enteredinto a multivariate logistic regression model, ensuring that theomnibus test of model coefficients was significant (P < .001). Allstatistical analysis was performed using SPSS version 19.0(SPSS Inc., Chicago, IL).

RESULTSOf the 499 patients, 45 (9.6%) developed FBS. The

mean age at FBS onset was 50 years (range, 12–81years), with a mean interval since surgery of 97 days(range, 6–877 months; median, 54 months). Average fol-low-up time was 39 months (range, 3–224 months) forall patients and 40 months (range, 3–149 months) forFBS patients. Patient characteristics and FBS incidenceare summarized in Table I. The mean age of all patientsin the study was 56 years (range, 2–91 years). On uni-variate analysis, patient factors significantly associatedwith the development of FBS were female gender (oddsratio [OR], 2.6; 95% confidence interval [CI], 1.4–5.1;P ! .003), absence of prior radiation (OR % P ! .023),and history of psychiatric illness (OR, 2.4; 95% CI, 1.2–5.1; P ! .015). Forty-seven patients had a prior historyof radiation to the head or neck (9%), none of whomdeveloped FBS.

Tumor characteristics and FBS incidence are sum-marized in Table II. Tumor factors significantlyassociated with the development of FBS on univariateanalysis were parotid site of origin (OR, 14.8; 95% CI,3.4–64; P < .001), schwannoma (OR, 6.9; 95% CI, 3.0–16;P < .001), and benign histology (OR, 33; 95% CI, 7.8–140; P < .001). Patients with tumors arising in the PPSand parotid gland accounted for all cases of FBS; nocases were observed in patients with tumors arising inthe ITF or other sites. The majority of parotid FBS cases(92%) were associated with tumors limited to the deeplobe. Most FBS cases (87%) were associated with threehistological classes: schwannoma (10; 22%), paragan-glioma (nine; 20%), or pleomorphic adenoma (20; 44%).Of 265 patients with resection of a malignant tumor,only two developed FBS (0.8%).

Surgical details and FBS incidence are summarizedin Table III. Surgical factors significantly associatedwith the development of FBS were transcervicalapproach (OR, 3.3; 95% CI, 1.8–6.3; P < .001), ITF dis-section (OR, 0.15; 95% CI, 0.04–0.63; P ! .003), PPSdissection (OR, 26; 95% CI, 7.9–85; P < .001), sympa-thetic chain sacrifice (OR, 9.2; 95% CI, 3.6–23; P < .001),resection involving the deep lobe of the parotid gland(OR, 0.53; 95% CI, 0.28–0.98; P ! .042), resection involv-ing the superficial lobe of the parotid gland (OR, 0.06;95% CI, 0.02–0.16; P < .001), and total parotid glandresection (OR, 0.05; 95% CI, 0.006–0.34; P < .001).

Postoperative findings are summarized in Table IV.Postoperative factors significantly associated with FBSwere Horner syndrome (OR, 12; 95% CI, 4.6–29; P <.001), temporomandibular joint pain (OR, 6.2; 95% CI,

1.8–22; P ! .001), facial nerve grades 1 to 3 (OR, 8.3;95% CI, 2.0–35; P ! .001), and adjuvant radiotherapy(OR, 0.06; 95% CI, 0.01–0.23; P < .001). Of 211 patientsreceiving adjuvant radiotherapy, only two developedFBS (1%).

Multivariate analysis was performed to identify var-iables independently associated with the development ofFBS; results are detailed in Table V. The statisticallysignificant factors were sympathetic chain sacrifice (OR,4.7; 95% CI, 1.5–15; P ! .008), PPS dissection (OR, 8.7;95% CI, 2.5–30; P ! .001), and extent of parotidectomy(resection of only deep lobe of parotid gland: OR, 40;

TABLE 3.Surgical Details and FBS Incidence.

Detail No. (%) Incidence (%) p

Approach

Transparotid 326 (65) 26 (8) <0.001

Transcervical 101 (20) 19 (19)

Anterior 72 (14) 0

Mandibulotomy

Yes 26 (5) 2 (8) 0.808

No 473 (95) 43 (10)

ITF Dissection

Yes 110 (22) 2 (2) 0.003

No 389 (78) 43 (11)

PPS Dissection

Yes 201 (40) 42 (21) <0.001

No 298 (60) 3 (1)

PPS Site

Pre-styloid 113 (56) 22 (19) 0.227

Post-styloid 62 (31) 17 (27)

Sympathetic Chain Sac

Yes 21 (4) 9 (43) <0.001

No 478 (96) 36 (8)

IJV Sac

Yes 34 (7) 0 0.057

No 465 (93) 45 (10)

ECA Ligation

Yes 38 (8) 6 (16) 0.129

No 461 (92) 39 (9)

Deep Parotid Resected

Yes 344 (69) 25 (7) 0.042

No 155 (31) 20 (13)

Superficial Parotid Resected

Yes 290 (58) 4 (1) <0.001

No 209 (42) 41 (20)

Total Parotid Resected

Yes 149 (30) 1 (1) <0.001

No 350 (70) 44 (13)

Gross Total Resection

Yes 489 (98) 45 (9) 0.315

No 10 (2) 0

Abbreviations: FBS, first bite syndrome; PPS, parapharyngeal space;ITF, infratemporal fossa; IJV, internal jugular vein; ECA, external carotidartery.

Laryngoscope 122: August 2012 Linkov et al.: First Bite Syndrome

1775

95% CI, 4.9–320; P ! .001; no parotidectomy performed:OR, 9.5; 95% CI, 1.2–78; P ! .036).

The cumulative incidence of FBS, calculated usingthe Kaplan–Meier method, was 9.6%. In patients under-going sympathetic chain sacrifice, the incidence was48.6% (P < .001); in patients undergoing PPS dissection,the incidence was 22.4% (P < .001). Whereas patientsundergoing only deep lobe of parotid gland resection hada 38.4% incidence of FBS, those undergoing total paroti-dectomy had an FBS incidence of 0.8% (P < .001). Thesefindings are presented in Figure 1. Patients whoreceived radiation either before or after definitive sur-gery had a 0.9% incidence of FBS, whereas patients whonever received radiation to the head or neck had a17.5% incidence of FBS (P < .001).

At last follow-up, FBS symptoms improved in 69%of patients (31 of 45). Fifteen patients with FBS (33%)received at least one form of treatment, including an an-algesic (n ! 9), an opioid (n ! 5), a neuropathic painmedication (n ! 10), acupuncture (n ! 1), and/or a Botoxinjection (n ! 1).

Structured telephone interview results are shownin Table VI. After attempting to reach every patientidentified with FBS, 16 of 45 were ultimately inter-viewed. Fourteen patients (88%) reported persistent FBSat time of interview; only two patients (12%) reported

TABLE 4.Postoperative Findings and FBS Incidence.

Finding No. (%) Incidence (%) p

FBS

Mean Age at Onset, y 50 (12-81)

Average Interval Since Surgery, d 97

Range Interval Since Surgery, d 6-877

FBS Treatment Received

Yes 15 (33)

No 30 (67)

FBS Improved

Yes 31 (69)

No 14 (31)

Horner’s Syndrome

Yes 21 (4) 10 (48) <0.001

No 478 (96) 35 (7)

TMJ Pain

Yes 11(2) 4 (36) 0.001

No 488 (98) 41 (8)

FN Fully Intact

Yes 206 (41) 23 (11) 0.160

No 293 (59) 22 (8)

FN Grade

Mean 2.5

1-3 371 (74) 43 (12) 0.001

4-6 128 (26) 2 (2)

Adjuvant RT

Yes 211 (42) 2 (1) <0.001

No 288 (58) 43 (15)

Follow Up Time (All Patients), m

Mean 39 (3-224)

Follow Up Time (FBS Patients), m

Mean 40 (3-149)

Abbreviations: FBS, first bite syndrome; TMJ, temporomandibularjoint; FN, facial nerve; RT, radiotherapy.

TABLE 5.Multivariate Analysis of Factors Associated with the

Development of FBS.

VariableNo.(%)

Incidence(%)

OR(95% CI) p

Gender

Male 261(52) 14 (5) Ref

Female 238 (48) 31 (13) 1.2 (0.56-2.7) 0.596

Prior RT

Yes 47 (9) 0 Ref

No 452 (91) 45 (10) 5.4E7 (0) 0.997

Psychiatric History

Yes 64 (13) 11 (17) 1.9 (0.76-4.7) 0.169

No 435 (87) 34 (8) Ref

ECA Ligation

Yes 38 (8) 6 (16) 1.7 (0.50-5.5) 0.412

No 461 (92) 39 (9) Ref

Sympathetic Chain Sac

Yes 21 (4) 9 (43) 4.7 (1.5-15) 0.008

No 478 (96) 36 (8) Ref

PPS Dissection

Yes 201 (40) 42 (21) 8.7 (2.5-30) 0.001

No 298 (60) 3 (1) Ref

Neuropathy

Yes 28 (6) 5 (18) 1.4 (0.36-5.3) 0.633

No 471 (94) 40 (8) Ref

Parotid Resected

No 153 (31) 20 (13) 9.5 (1.2-78) 0.036

Superficial Lobe Only 2 (.4) 0 0 1.0

Deep Lobe Only 56 (11) 21 (38) 40 (4.9-317) 0.001

Subtotal 139 (28) 3 (2) 3.7 (0.37-37) 0.266

Total 149 (30) 1 (1) Ref

Abbreviations: FBS, first bite syndrome; RT, radiotherapy; ECA,external carotid artery; Sac, sacrifice; PPS, parapharyngeal space.

Fig. 1. Incidence of first bite syndrome (FBS) for risk factors inde-pendently significant on multivariate analysis. OR ! odds ratio.

Laryngoscope 122: August 2012 Linkov et al.: First Bite Syndrome

1776

complete resolution of symptoms. On a pain scale from 1to 10, with 1 indicating occasional low-grade pain and10 representing severe, constant pain with every meal,the average intensity of symptoms was 5.7 (range, 1–10). Patients described the pain as sharp, shooting (fourpatients), worse with sour foods (seven patients), andworse after long breaks between meals (six patients).Five patients (31%) received some form of treatment,with variable efficacy. Twelve patients (75%) reportedmodifying their eating behavior to adjust to FBS. Adapt-ive strategies included mentally preparing/closing theeyes (n ! 3), chewing on the unaffected side (n ! 3),

clenching fist/banging foot (n ! 2), taking small bites offood (n ! 2), rubbing the parotid area with onset of pain(n ! 2), and avoiding sour foods (n ! 2). Only twopatients recalled being counseled preoperatively on therisk of developing FBS. However, no patient indicated heor she would have declined surgery, had he or sheknown FBS would develop. Preoperative consent discus-sions from all 45 FBS patients were reviewed formention of FBS. Whereas only 3 patients were explicitlytold of the risk of developing FBS, 17 were told of thepotential for developing dysfunction of swallowing.

DISCUSSIONOur study goals were to identify the incidence, risk

factors, treatment, and outcomes of FBS. We identified acomprehensive cohort of patients undergoing surgicalprocedures in anatomic locations associated with the de-velopment of FBS. These locations were the PPS, ITF(containing the middle meningeal artery), and deep lobeof the parotid gland. Of 499 patients, 45 developed FBSpostoperatively, for a cumulative incidence of 9.6%. Theincidence of FBS was particularly high in patientsundergoing PPS dissection (22.4%), resection of only thedeep lobe of the parotid gland (38.4%), and sacrifice ofthe sympathetic chain (48.6%). These three variableswere significant independent predictors of FBS on multi-variate analysis.

The term first bite syndrome was used in 1998 byNetterville et al., referring to facial pain characterizedby a severe cramping or spasm in the parotid regionwith the first bite of each meal that diminishes over thenext several bites. FBS was initially attributed tosympathetic denervation. In a case series of 46 patientswith vagal paraganglioma, the authors proposed a mech-anism of loss of sympathetic innervation to the parotidgland causing a denervation supersensitivity thatwas activated by parasympathetic neurotransmitters,particularly during initial oral intake. Parasympathetichyperactivation is speculated to stimulate an exagger-ated myoepithelial cell contraction throughout theparotid gland, causing pain. In support of this mecha-nism, eight of the nine FBS cases in the Nettervilleet al. series underwent sympathetic trunk resection, andthe other patient exhibited sympathetic dysfunctionpostoperatively as manifested by Horner syndrome.1,2,4

TABLE 6.Results of FBS Phone Questionnaire (n516) and Consent Data.

Item No. (%)

Symptoms Present at Interview 14 (88)

Average Intensity of Symptoms (1–10) 5.7

Quality of Symptoms

Worse with sour food 7 (44)

Worse after long food breaks 6 (38)

Sharp, shooting pain 4 (25)

Treated for FBS 5 (31)

Eating Behavior Modified Because of FBS 12 (75)

Type of Behavior Modification

Mentally Prepare/Close Eyes 3

Chew on Other Side 3

Clench Fist/Bang Foot 2

Small Bites of Food 2

Rub Parotid Area 2

Avoid Sour Foods 2

Recall being told preoperatively about risk of FBS 2 (13)

Would Have Liked to Know About FBS Before Surgery 9 (56)

If you knew about the risk of FBS, would you have

reconsidered having surgery?

Yes 0

No 16 (100)

Consent Data (n ! 45)

Told of potential for ‘‘dysfunction of speechand swallowing’’

17 (38)

Told of potential for ‘‘first bite syndrome’’ 3 (7)

Abbreviation: FBS, first bite syndrome.

TABLE 7.Summary of Literature Reported Case Series with FBS Development (min. >1 case).

Study Site(s) Assessed Total Cases

FBS Cases(% incidence

of FBS)

For FBS Cases (% of total FBS cases)

PPSinvolvement

SympatheticChain Sacrifice

ECALigation

Linkov et al. 2011 ITF, PPS, DP 499 45 (9.6) 42 (93) 9 (20) 6 (13)

Chijiwa et al. 2009 PPS 24 7 (29) 7 (100) N/A N/A

Kawashima 2008 PPS 22 9 (41) 9 (100) 5 (56) 6 (67)

Chiu et al. 2002 PPS 12 12 (100) 12 (100) 5 (42) 6 (50)

Netterville et al. 1998 PPS 46 9 (20) 9 (100) 8 (89) N/A

Abbreviations: FBS, first bite syndrome; ITF, infratemporal fossa; PPS, parapharyngeal space; DP, deep lobe of parotid; ECA, external carotid artery; N/A, not available.

Laryngoscope 122: August 2012 Linkov et al.: First Bite Syndrome

1777

Other case series have supported the role of sympa-thetic denervation in the development of FBS. In 2002,Chiu et al. published a series of 12 patients with FBS af-ter surgery involving the PPS. Only six patients hadtheir sympathetic chain transected and/or exhibitedHorner syndrome postoperatively. The remaining sixpatients had undergone ligation of their external carotidartery (ECA). The authors revisited salivary gland phys-iology and anatomy to explain that all 12 patients likelyexperienced a loss of sympathetic innervation to the pa-rotid gland, supporting the initial theory by Nettervilleet al.4

In 2008, Kawashima et al. reviewed 22 patientswith PPS tumors and identified nine cases of FBS, eightof which involved surgical ablation of the cervical sym-pathetic chain and/or ECA.5 Further adding support tothe sympathetic denervation hypothesis, there haverecently been several reports of patients exhibiting FBSpreoperatively due to tumor invasion of the sympatheticchain and/or ECA.7–9 A summary of the published caseseries (involving >1 case per study) with FBS develop-ment, including the results of the present study, can befound in Table VII. In the three small case series assess-ing only PPS tumors, the incidence of FBS ranges from20% to 41%.1,5,6 Our results fall within this range forPPS dissection and deep parotid lobe resection, but pointto a higher incidence when the sympathetic trunk isresected.

In addition to the three variables found to be signif-icant independent predictors of FBS on multivariateanalysis (PPS dissection, deep parotid lobe resection,and sympathetic trunk ablation), several other notablevariables associated with FBS were identified via uni-variate analysis. An inverse relationship was notedbetween receiving radiation therapy to the head and/orneck before or after surgery and developing FBS. It isunclear whether the actual radiation leads to a reduc-tion in FBS incidence (perhaps due to a reduction infunctioning parotid gland myoepithelial cells) or if thephenomenon is a byproduct of other related variables,such as more benign disease in the nonradiated group.

Another variable of interest is the complete resec-tion of the parotid gland. In the present study, only onecase of FBS occurred when a total parotidectomy wasperformed. Because the pain in FBS is likely generatedby enhanced contractility of the myoepithelial cells inparotid tissue,4 it is expected that removal of all ipsilat-eral parotid tissue would eliminate FBS. This one

unusual case may have been associated with inadver-tently retained, or ectopically present, parotid tissue. Itappears that preservation of the superficial lobe whenperforming an isolated deep lobe parotidectomyincreases the risk of developing FBS.

Several treatment strategies were attempted forsome of the patients with FBS, including pain medica-tions, acupuncture, and Botox injections. Most patientswith and without treatment improved over time; how-ever, few experienced a complete resolution ofsymptoms. Although a thorough assessment of theeffects of specific treatment types on symptomatologywas not performed, there did not appear to be a consis-tently effective treatment option. As gleaned fromtelephone interviews, patients had tried various combi-nations of treatments, including medicinal, procedural,and adaptive behavioral techniques, to help cope withFBS.

CONCLUSIONHead and neck surgeons planning to operate in the

PPS or to resect the deep lobe of the parotid should beaware of the potential consequence of FBS and shouldcounsel their patients on its risk and anticipatedoutcome.

BIBLIOGRAPHY1. Netterville JL, Jackson CG, Miller FR, Wanamaker JR, Glasscock ME.

Vagal paraganglioma: a review of 46 patients treated during a 20-yearperiod. Arch Otolaryngol Head Neck Surg 1998;124:1133–1140.

2. Netterville JL, Jackson CG, Wanamaker JR, Miller FR, Glasscock ME.Va-gal paraganglioma: a review of 44 patients treated over 18 years. Pre-sented at the 40th Annual Meeting of American Society for Head andNeck Surgery/Society of Head and Neck Surgeons at Combined Otolar-yngology Spring Meetings; Palm Beach, Florida, U.S.A.; May 14–16,1998.

3. Haubrich WS. The first-bite syndrome. Henry Ford Hosp Med J 1986;34:275–278.

4. Chiu AG, Cohen JI, Burningham AR, Andersen PE, Davidson BJ. First bitesyndrome: a complication of surgery involving the parapharyngealspace. Head Neck 2002;24:996–999.

5. Kawashima Y, Sumi T, Sugimoto T, Kishimoto S. First-bite syndrome: areview of 29 patients with parapharyngeal space tumor. Auris NasusLarynx 2008;35:109–113.

6. Chijiwa H, Mihoki T, Shin B, Sakamoto K, Umeno H, Nakashima T. Clini-cal study of parapharyngeal space tumours. J Laryngol Otol 2009;123(suppl 31):100–103.

7. Lieberman SM, Har-El G. First bite syndrome as a presenting symptom ofa parapharyngeal space malignancy. Head Neck 2011;33:1539–1541.

8. Deganello A, Meccariello G, Busoni M, Franchi A, Gallo O. First bite syn-drome as presenting symptom of parapharyngeal adenoid cystic carci-noma. J Laryngol Otol 2011;125:428–431.

9. Diercks GR, Rosow DE, Prasad M, Kuhel WI. A case of preoperative ‘‘first-bite syndrome’’ associated with mucoepidermoid carcinoma of the pa-rotid gland. Laryngoscope 2011;121:760–762.

Laryngoscope 122: August 2012 Linkov et al.: First Bite Syndrome

1778