The Laryngoscope C 2015 The American Laryngological, V
Rhinological and Otological Society, Inc.
Percutaneous Endoscopic Gastrostomy Tube Dependence Following Chemoradiation in Head and Neck Cancer Patients Surjeet Pohar, MD; Michael Demarcantonio, MD; Phillip Whiting, MD; Edwin Crandley, MD; John Wadsworth, MD; Daniel Karakla, MD Objectives/Hypothesis: Compare long-term percutaneous endoscopic gastrostomy (PEG) tube dependence, stricture rate, and weight loss in patients receiving a prophylactic gastrostomy tube with those who initially rely on oral intake during chemoradiation for head and neck cancer. Also, to determine what other patient and treatment characteristics influence development of long-term severe dysphagia. Study Design: Retrospective review. Methods: Seventy-nine patients received a PEG tube and 25 did not. The prophylactic and initial oral intake groups were then analyzed to assess the primary outcomes of PEG dependence at last follow-up >1 year. Results: On univariate and multivariate analysis, Zubrod score >1, prophylactic PEG placement, and higher T classification were predicted for PEG tube dependence at last follow-up at least 1 year after treatment. Conclusions: Prophylactic PEG tube, high Zubrod score, and high T stage were independent predictors for PEG tube dependence at least 1 year after treatment in patients with head and neck cancer receiving definitive chemoradiation. Key Words: Radiation therapy, swallowing, dysphagia. Level of Evidence: 4 Laryngoscope, 00:000–000, 2015
INTRODUCTION In recent years, organ-sparing chemoradiation therapy (CRT) has assumed an ever-expanding role in the treatment of head and neck cancer. CRT has demonstrated acceptable locoregional control and survival, but this therapy often results in significant complications. A major and often-cited complication is that of dysphagia. Head and neck cancer post-treatment dysphagia occurs in 50% to 60% of patients or 10,000 to 20,000 new cases per year.1 Severe dysphagia can combine with xerostomia and mucositis to have a significant effect on a patient’s oral intake.2 To maintain adequate nutrition during CRT, a patient’s oral intake often requires nonoral nutritional supplementation. Traditionally, nutritional supplementation has been accomplished via either a nasogastric (NG) or percutaneous endoscopic gastrostomy (PEG) tube. PEG tube placement has emerged as the primary means of supplementing nutrition during From the Department of Radiation Oncology (S.P., E.C.), Department of Otolaryngology–Head and Neck Surgery (M.D., P.W., D.K.), Eastern Virginia Medical School, Norfolk, Virginia; and the Department of Otolaryngology–Head and Neck Surgery (J.W.), Emory University, Atlanta, Georgia Editor’s Note: This Manuscript was accepted for publication December 4, 2014. Presented at the American Academy of Otolaryngology–Head and Neck Surgery Annual Meeting, Orlando, Florida, U.S.A., September 21– 24, 2014. The authors have no funding, financial relationships, or conflicts of interest to disclose. Send correspondence to Surjeet Pohar, MD, Indiana University Health East, 6845 Rama Drive, Indianapolis, IN 46219. E-mail: spohar@ iuhealth.org DOI: 10.1002/lary.25117
Laryngoscope 00: Month 2015
treatment. PEG placement has been shown to have minimal complications and to effectively limit weight loss.2 In a survey of head and neck cancer patients, Terrell et al. demonstrated that continued feeding tube use after treatment had the greatest negative effect on quality of life of any factor analyzed, including total laryngectomy and more than two comorbid conditions.3 We decided to use our patient data to determine whether PEG tube placement versus reliance on oral intake prior to CRT affected long-term severe dysphagia and weight loss during CRT. We also wanted to determine what the stricture rate was in our patients and whether there was a correlation between stricture development and long-term severe dysphagia.
MATERIALS AND METHODS The Eastern Virginia Medical School Cancer registry was queried to identify all patients with a new head and neck cancer diagnosis from January 1, 2001 to July 1, 2008. A total of 223 patients with stage III or IV squamous cell head and neck (SCCHN) carcinoma were identified. After exclusion of patients treated with primary surgical therapy, those receiving CRT at an outside facility, those requiring PEG tube feeding prior to therapy, those treated with radiation therapy (XRT) only, and those lost to follow-up, a total of 129 patients were included. All patients were reviewed for demographic information including age, primary site, stage, and Zubrod score at presentation. Patients were classified by Zubrod into those with Zubrod >1 (unable to work) or Zubrod of 0 or 1 (able to do at least light work). Ninety-nine patients had a prophylactic PEG tube placed (PPEG group), whereas 30 did not (initial oral intake group). Of this group, 22 died of disease within 1 year, two died of other causes, and one had no follow-up swallowing status reported.
Pohar et al.: PEG Tube Dependence Following Chemoradiation
1
TABLE I. Radiation Therapy Details in Patients with >1-Year Follow-up. Prophylactic, n 5 79
No Prophylactic PEG, n 5 25
TABLE III. Patient Characteristics for Those With >1-Year Follow-up (n 5 104). Prophylactic PEG,No. (%)
PO, No. (%)
Age, yr, median
57
53
Follow up (months), median
36
48
57 (72)
22 (28)
22 (88)
3 (12)
P
Technique Standard, n (%) IMRT, n (%)
69 (66) 10 (10)
20 (19) 5 (5)
Median radiation dose, cGy
7,030
7,030
Treatment delay, median, d
3
4
.35
Sex Male
IMRT 5 intensity-modulated radiation therapy; PEG 5 percutaneous endoscopic gastrostomy.
Thus, 104 patients were alive, had at least 12 months of followup, and had swallowing status recorded at the last follow-up. Swallowing status at the last follow-up was classified as PEG dependent or diet via oral intake. In this group of 104 patients, 79 had a PPEG and 25 began therapy with only oral intake. All patients underwent evaluation by a head and neck surgeon followed by a staging endoscopy. The need for gastrostomy tube placement was made on an individual basis based on tumor site, patient condition, and patient wishes. Sixty-three of 70 patients (90%) managed by one surgeon had a prophylactic PEG tube placed versus 36 of 59 (61%) managed by the other surgeon (P 5.002). All patients then underwent definitive treatment consisting of external beam radiation therapy with concurrent chemotherapy at Sentara Norfolk General Hospital or an affiliated facility. Patients were not given any specific instruction on swallowing exercises; however, most patients did see a dietitian who encouraged adequate nutritional intake either orally or by tube feeding. Patients in both groups were encouraged to continue oral intake. Most patients underwent standard two-dimensional or three-dimensional conformal radiation therapy. A smaller proportion of patients were treated with intensity-modulated radiation therapy (IMRT) (Table I). Various chemotherapy regimens were used, with weekly carboplatin/taxol being the most common (Table II). All patients were followed by radiation oncology, hematology-oncology, and otolaryngology for complaints of dysphagia, decreased oral intake, weight loss, and hospital admission. Those who did not have a prophylactic PEG (initial oral intake group) tube, had a therapeutic PEG tube placed if they had inadequate or no oral intake during treatment. The patients’ clinical courses were followed to assess the primary outcomes of swallowing/diet at 6 months, 1 year, and at last follow-up, as well as the need for esophageal dilation and treatment delay. Treatment delay was calculated by estimating expected date of completion of XRT treatment. Additional outcomes of local and regional recurrence as well as survival were
Female Race African American Caucasian
23 (85)
4 (15)
56 (73)
21 (27)
Overall stage Stage III
21 (27)
8 (32)
Stage IV
58 (73)
17(68)
T stage T0
1 (1)
3 (12)
T1
11 (14)
3 (12)
T2 T3
17 (22) 27 (34)
8 (32) 6 (24)
T4
23 (29)
5 (20)
N stage N0
16 (20)
6 (24)
N1
14 (18)
4 (16)
N2 N3
40 (51) 9 (11)
11 (44) 4 (16)
Primary site larynx/hypopharynx
24 (30)
3 (12)
Other Zubrod
55 (70)
22(88)
0–1
65 (86)
21 (88)
>1
11 (14)
3 (12)
P Value
.11
.19
.60
.09
.87
.07
.81
PEG 5 percutaneous endoscopic gastrostomy; PO 5 oral intake.
also collected. All aspects of the research were approved by the Eastern Virginia Medical School Institutional Review Board. SPSS statistical package version 18.0 was used for analysis (SPSS Inc., Chicago, IL).
RESULTS Patient Demographics
TABLE II. Chemotherapy in 104 Patients With >1-year Follow-up. Chemotherapy Regimen
Cisplatin weekly High-dose cisplatin every 3 weeks
No. of Patients
14 2
Other cisplatin-only regimen
2
Carboplatin/taxol weekly Cetuximab weekly
52 2
Carboplatin weekly
2
Other combination
27
Laryngoscope 00: Month 2015
2
Patient characteristics are described in Table III. Overall, out of 104 patients who had at least 1 year of follow-up, 79 patients underwent prophylactic PEG tube placement at the time of pan-endoscopy (PPEG group). An additional 25 patients received no feeding conduit at the time of staging (initial oral intake group). Of these 25 patients, nine required gastrostomy tube placement during CRT, and two patients had tubes placed after CRT (therapeutic PEG). No significant differences between the PPEG and initial oral intake groups with regard to age, sex, race, Zubrod score, T stage, N stage, and overall stage (Table III) were found. Median followup for these 104 patients was 39 months. Pohar et al.: PEG Tube Dependence Following Chemoradiation
TABLE IV. Five-Year Kaplan-Meier Adjusted Control and Survival. Prophylactic PEG
No Prophylactic PEG
P value
Local control
84.8%
80.0%
0.58
Regional control
88.9%
93.3%
0.73
Overall survival Disease specific survival
55.6% 65.5%
73.3% 78.6%
0.09 0.24
PEG 5 percutaneous endoscopic gastrostomy.
Hypopharyngeal or laryngeal carcinoma patients were more likely to receive a PPEG than other primary site patients (P 5.07).
Survival In comparing the survival for all 129 patients in the prophylactic PEG versus initial oral intake groups, no difference was found in terms of 5-year estimates of local control, regional control, overall survival, and disease-specific survival (Table IV).
Radiation Therapy No significant differences were observed in regard to radiation method, overall dose, or treatment delay for the 104 patients with at least 1-year follow-up (Table I).
Swallowing The surrogate endpoint used for poor swallowing ability was PEG tube dependence. There was no difference in PEG tube dependence between the PPEG and oral intake groups at 6 months (P 5.27). However, PEG tube dependence at last follow-up at least 1 year out
from therapy was 4% in those with no initial PEG versus 23% with prophylactic PEG (P 5.03). Other factors associated with PEG tube dependence in univariate analysis were poor functional status defined by a Zubrod score >1 and higher T stage (Table V). Other factors examined, including stage, N stage, and smoking status, were not significant (Table V). Statistically significant factors and subsite were included in a multivariate analysis of the 104 patients with more than 1-year follow-up. This demonstrated that PEG tube dependence was associated with higher T stage, Zubrod score >1, and use of prophylactic PEG feeding (Table V). Four patients had dysphagia prior to CRT, and all had a PPEG (one did not have a 1-year follow-up). Excluding the three with >1-year follow-up from the univariate and multivariate analysis did not change our findings (Table V). The raw data for multivariate analysis of the diet at last followup versus prognostic factor are shown in Table VI. The need for esophageal dilation is an important complication and comorbidity of CRT. Twenty-five of the 99 patients (25.3%) receiving a prophylactic gastrostomy tube required esophageal dilation compared to four patients (13.3%) in the initial oral intake group (P 5.22). There was no correlation between development of stricture requiring dilation and PEG dependence at last follow-up (n 5 129, P 5.65), but there was a trend toward a correlation when only including patients with >1-year follow-up (n 5 104, P 5.11). Data on weight loss during treatment for the entire group of 129 patients studied were available in 82 patients who had a prophylactic PEG tube and 26 patients who did not. There was no significant difference in the average weight decline of 8 kg (range, 13.8 kg to 227.7 kg) from 78 kg to 70 kg in patients with a prophylactic PEG tube and 8 kg from 90 kg to 82 kg (range, 12.2 to 217.1 kg) in those relying initially on oral intake (P 5.95).
TABLE V. Univariate and Multivariate Analysis of Prognostic Factors Versus PEG Dependence at Last Follow-up >1 Year (n 5 104), Excluding Patients With Pretreatment Dysphagia (n 5 101). Univariate, n 5 104, P Value
Multivariate, n 5 104, P Value
Univariate Excluding Dysphagia, n 5 101, P Value
Multivariate Excluding Dysphagia, n 5 101, P Value
Age Sex
.60 .40
— —
.57 .31
— —
Race (African American vs. white)
.24
—
.27
—
T stage N stage
.008 .29
.01 —
.009 .25
.017
Stage (III vs. IV)
.13
—
.09
—
Smoking Alcohol use
.60 .63
— —
.60 .63
— —
Variable
Initial PEG vs. no PEG Zubrod (>1 vs. 0–1) Laryngeal/hypopharyngeal primary
.03
.04
.03
.04
.000001 0.97
.000001 0.10
.000002 0.86
.000001 0.13
PEG 5 percutaneous endoscopic gastrostomy.
Laryngoscope 00: Month 2015
Pohar et al.: PEG Tube Dependence Following Chemoradiation
3
TABLE VI. Multivariate Analysis Raw Data for Swallowing at Last Follow-up >1 Year (n 5 104). Swallowing, No. PEG Dependent
PO
Initial PEG
18
61
No initial PEG Zubrod score
1
24
10
76
9
5
0
0
4
1 2
0 2
14 23
3
10
23
7
21
Laryngeal/hypopharyngeal
5
22
Other primary site
14
63
0–1 >1 T stage
4 Primary site
P-value
0.04
0.000001 0.01
0.10
PEG 5 percutaneous endoscopic gastrostomy; PO 5 oral intake.
DISCUSSION Weight Loss and Swallowing The need for prophylactic PEG placement in CRT remains a controversial topic. It is undeniable that CRT results in significant morbidity, with up to 86% of patients experiencing grade 3 or 4 mucositis.2 As a result, some authors have recommended universal placement of percutaneous gastrostomy tubes in all head and neck cancer patients prior to CRT.2 The primary argument for this approach is the prevention of malnutrition, weight loss, and subsequent hospitalization. Lee et al. examined the effectiveness of prophylactic PEG tube placement in 88 patients with advanced head and neck cancer.2 Of their cohort, 41% of patients received a gastrostomy tube prior to beginning CRT or twice-daily radiation alone. Those patients not receiving a gastrostomy tube lost an average of 7.0 kg versus 3.1 kg in the PPEG group (P 1 year from treatment in patients who initially had only oral intake. Several authors have tried to determine what treatment or patient attributes are associated with long-term dysphagia in CRT patients. Eisbruch et al. have found a correlation between radiation dose to the pharyngeal constrictors and long-term dysphagia.6 Caudell et al. and others have shown that laryngeal/hypopharyngeal cancer patients are more likely than other head and neck cancer patients to have long-term swallowing difficulties.7 They also found that a median dose >50 Gy to the larynx or inferior pharyngeal constrictor correlated with a higher risk of long-term severe dysphagia. When they excluded laryngeal/hypopharyngeal patients, they found that higher doses to the middle and superior constrictor also correlated with increased risk of developing long-term severe dysphagia.8 Sanguineti et al. found that higher oral cavity dose correlates with higher longterm dysphagia rates.9 Chen et al. found that 1-year PEG tube dependence was 21% in their PPEG patients versus zero in the initial oral intake group (P 1, and higher T stage to predict for PEG tube dependence at last follow-up >1 year. A summary of various series reporting on swallowing dysfunction, stricture, and weight loss is presented in Table VII. Our patients did not have routine consultation with a speech pathologist to initiate a program of prophylactic swallowing exercises to prevent dysphagia. Most studies of prophylactic swallowing exercises suggest an improvement in dysphagia; however, they generally report only 3 to 6 months follow-up and have small datasets.11,12 Hutcheson et al. recently reported on 497 patients treated with XRT alone or CRT for head and neck cancer.13 They found on multivariate analysis that maintaining swallowing exercises or a full diet throughout therapy each predicted for a higher likelihood of maintaining long-term regular dietary intake. Our current practice is to support patients with analgesics and counsel them to maintain oral intake even if they have a PEG tube, and to refer to speech pathology at the time of diagnosis for formal swallow evaluation and initiation of prophylactic swallowing exercises during treatment. The findings of the randomized trial by Silander et al. of a lower rate of long-term PEG tube dependence in the PPEG group is the opposite of those in our study and in Chen et al.’s data.4 Further studies are needed to see if our findings are confirmed in patients treated with CRT using IMRT only. We found that Zubrod score >1 (unable to work) correlated with long-term PEG tube dependence. We are not sure why this is, but perhaps patients with more functional disability also have less ability to overcome tissue injury from CRT. Higher T-stage patients may have had higher rates of long-term swallowing dysfunction because they had a larger volume of tissue treated to a high dose, resulting in more tissue being damaged beyond recovery (Table VI).
Stricture We found a stricture rate requiring dilation of 25% in our PPEG group, similar to the 30% with PPEG reported by Chen et al.10 Clavel et al. and Mekhail et al., using a policy of initial oral intake with a therapeutic NG tube as needed, found the rate of stricture to be only 3% and 4%, respectively.5,14 They hypothesized that the NG tube served a stenting function. However, Chen et al. found a 6% stricture rate, and we found a Laryngoscope 00: Month 2015
13% rate for patients on initial oral intake during CRT with therapeutic PEG placement as needed.10 Furthermore, NG tubes are uncomfortable for patients to have for the median of 40 days in the Clavel study and 8 weeks in Mekhail et al.’s group.5,14 Mekhail et al. found that 23% of therapeutic PEG tube patients required pharyngoesophageal dilation compared with only 4% of patients treated with a therapeutic NG feeding tube during CRT.14 The authors did state that patients selected for PEG tube placement were more malnourished than those treated with NG tubes. A summary of complications in the various series is presented in Table VII.
Treatment Delay The effect of treatment delay on overall survival was examined by Fesinmeyer et al. by reviewing Medicare data from 3,864 head and neck cancer patients.15 Their research demonstrated that in laryngeal cancer, a treatment delay of 5 to 30 days resulted in a 68% increased risk of death compared with patients with no interruptions. Some have thought prophylactic PEG would reduce treatment delays and thus improve effectiveness of CRT.2 Our results, similar to those of Chen et al.,10 failed to find a significant difference in treatment delay or locoregional control between the PPEG group and oral intake group. Considering these findings, it is questionable whether universal PEG tube placement is necessary or beneficial in regard to treatment delays during chemoradiation. When evaluating the need for prophylactic PEG tube placement, one must consider that placement is not entirely benign. Although PEG tube insertion presents a small risk of complication, the risk is real. Complications such as colon perforation can occur. In addition to immediate comorbid effects, some researchers have demonstrated increased esophageal toxicity in those patients undergoing PEG tube placement.10,14 It is important to acknowledge that as with most research on this topic, our collection method was retrospective with obvious limitations. These include imbalance between number of patients having a PPEG and initial oral intake, and selection bias so that a larger proportion of high–T-stage patients and hypopharyngeal/ laryngeal patients received PPEG (Table III). Ideally, this would be corrected by completing a randomized trial stratifying patients for known prognostic factors from our study and other retrospective studies. Also, such a study would use current radiation techniques of IMRT and image-guided radiation therapy. However, the only randomized trial comparing prophylactic PEG tube versus oral intake in exclusively CRT patients was closed because of low patient accrual.16
CONCLUSION Patients with head and neck carcinoma treated with CRT were found to have higher rates of long-term PEG tube dependence if they had a PPEG, had a higher T stage, or had a Zubrod score >1. Prophylactic PEG tube placement did not reduce treatment breaks during the radiation course, affect overall survival, or reduce weight Pohar et al.: PEG Tube Dependence Following Chemoradiation
5
loss during CRT. Stricture rates were lower in patients started initially on oral intake than with prophylactic PEG. It seems that initial oral intake with therapeutic PEG as needed is the better approach. Clearly, further prospective research is needed to develop standards and protocols for the placement of prophylactic PEG placement.
BIBLIOGRAPHY 1. Kulbersh BD, Rosenthal EL, McGrew BM, et al. Pretreatment, preoperative swallowing exercises may improve dysphagia quality of life. Laryngoscope 2006;116:883–886. 2. Lee JH, Machtay M, Unger LD, et al. Prophylactic gastrostomy tubes in patients undergoing intensive irradiation for cancer of the head and neck. Arch Otolaryngol Head Neck Surg 1998;124:871–875. 3. Terrell JE, Ronis DL, Fowler KE, et al. Clinical predictors of quality of life in patients with head and neck cancer. Arch Otolaryngol Head Neck Surg 2004;130:401–408. 4. Silander E, Nyman J, Bove M, Johansson L, Larsson S, Hammerlid E. Impact of prophylactic percutaneous endoscopic gastrostomy on malnutrition and quality of life in patients with head and neck cancer: a randomized study. Head Neck 2012;34:1–9. 5. Clavel S, Fortin B, Despres P, et al. Enteral feeding during chemoradiotherapy for advanced head-and-neck cancer: a single-institution experience using a reactive approach. Int J Radiat Oncol Biol Phys 2011;79:763–769. 6. Eisbruch A, Kim HM, Feng FY, et al. Chemo-IMRT of oropharyngeal cancer aiming to reduce dysphagia: swallowing organs late complication probabilities and dosimetric correlates. Int J Radiat Oncol Biol Phys 2011;81:e93–e99. 7. Caudell JJ, Schaner PE, Meredith RF, et al. Factors associated with longterm dysphagia after definitive radiotherapy for locally advanced headand-neck cancer. Int J Radiat Oncol Biol Phys 2009;73:410–415.
Laryngoscope 00: Month 2015
6
8. Caudell JJ, Schaner PE, Desmond RA, Meredith RF, Spencer SA, Bonner JA. Dosimetric factors associated with long-term dysphagia after definitive radiotherapy for squamous cell carcinoma of the head and neck. Int J Radiat Oncol Biol Phys 2010;76:403–409. 9. Sanguineti G, Gunn GB, Parker BC, Endres EJ, Zeng J, Fiorino C. Weekly dose-volume parameters of mucosa and constrictor muscles predict the use of percutaneous endoscopic gastrostomy during exclusive intensitymodulated radiotherapy for oropharyngeal cancer. Int J Radiat Oncol Biol Phys 2011;79:52–59. 10. Chen AM, Li BQ, Lau DH, et al. Evaluating the role of prophylactic gastrostomy tube placement prior to definitive chemoradiotherapy for head and neck cancer. Int J Radiat Oncol Biol Phys 2010;78:1026–1032. 11. Carnaby-Mann G, Crary MA, Schmalfuss I, Amdur R. "Pharyngocise": randomized controlled trial of preventative exercises to maintain muscle structure and swallowing function during head-and-neck chemoradiotherapy. Int J Radiat Oncol Biol Phys 2012;83:210–219. 12. van der Molen L, van Rossum MA, Burkhead LM, Smeele LE, Rasch CR, Hilgers FJ. A randomized preventive rehabilitation trial in advanced head and neck cancer patients treated with chemoradiotherapy: feasibility, compliance, and short-term effects. Dysphagia 2011;26:155–170. 13. Hutcheson KA, Bhayani MK, Beadle BM, et al. Eat and exercise during radiotherapy or chemoradiotherapy for pharyngeal cancers: use it or lose it. JAMA Otolaryngol Head Neck Surg 2013;139:1127–1134. 14. Mekhail TM, Adelstein DJ, Rybicki LA, Larto MA, Saxton JP, Lavertu P. Enteral nutrition during the treatment of head and neck carcinoma: is a percutaneous endoscopic gastrostomy tube preferable to a nasogastric tube? Cancer 2001;91:1785–1790. 15. Fesinmeyer MD, Mehta V, Blough D, Tock L, Ramsey SD. Effect of radiotherapy interruptions on survival in medicare enrollees with local and regional head-and-neck cancer. Int J Radiat Oncol Biol Phys 2010;78: 675–681. 16. Corry J, Poon W, McPhee N, et al. Prospective study of percutaneous endoscopic gastrostomy tubes versus nasogastric tubes for enteral feeding in patients with head and neck cancer undergoing (chemo)radiation. Head Neck 2009;31:867–876.
Pohar et al.: PEG Tube Dependence Following Chemoradiation
Rhinological and Otological Society, Inc.
Percutaneous Endoscopic Gastrostomy Tube Dependence Following Chemoradiation in Head and Neck Cancer Patients Surjeet Pohar, MD; Michael Demarcantonio, MD; Phillip Whiting, MD; Edwin Crandley, MD; John Wadsworth, MD; Daniel Karakla, MD Objectives/Hypothesis: Compare long-term percutaneous endoscopic gastrostomy (PEG) tube dependence, stricture rate, and weight loss in patients receiving a prophylactic gastrostomy tube with those who initially rely on oral intake during chemoradiation for head and neck cancer. Also, to determine what other patient and treatment characteristics influence development of long-term severe dysphagia. Study Design: Retrospective review. Methods: Seventy-nine patients received a PEG tube and 25 did not. The prophylactic and initial oral intake groups were then analyzed to assess the primary outcomes of PEG dependence at last follow-up >1 year. Results: On univariate and multivariate analysis, Zubrod score >1, prophylactic PEG placement, and higher T classification were predicted for PEG tube dependence at last follow-up at least 1 year after treatment. Conclusions: Prophylactic PEG tube, high Zubrod score, and high T stage were independent predictors for PEG tube dependence at least 1 year after treatment in patients with head and neck cancer receiving definitive chemoradiation. Key Words: Radiation therapy, swallowing, dysphagia. Level of Evidence: 4 Laryngoscope, 00:000–000, 2015
INTRODUCTION In recent years, organ-sparing chemoradiation therapy (CRT) has assumed an ever-expanding role in the treatment of head and neck cancer. CRT has demonstrated acceptable locoregional control and survival, but this therapy often results in significant complications. A major and often-cited complication is that of dysphagia. Head and neck cancer post-treatment dysphagia occurs in 50% to 60% of patients or 10,000 to 20,000 new cases per year.1 Severe dysphagia can combine with xerostomia and mucositis to have a significant effect on a patient’s oral intake.2 To maintain adequate nutrition during CRT, a patient’s oral intake often requires nonoral nutritional supplementation. Traditionally, nutritional supplementation has been accomplished via either a nasogastric (NG) or percutaneous endoscopic gastrostomy (PEG) tube. PEG tube placement has emerged as the primary means of supplementing nutrition during From the Department of Radiation Oncology (S.P., E.C.), Department of Otolaryngology–Head and Neck Surgery (M.D., P.W., D.K.), Eastern Virginia Medical School, Norfolk, Virginia; and the Department of Otolaryngology–Head and Neck Surgery (J.W.), Emory University, Atlanta, Georgia Editor’s Note: This Manuscript was accepted for publication December 4, 2014. Presented at the American Academy of Otolaryngology–Head and Neck Surgery Annual Meeting, Orlando, Florida, U.S.A., September 21– 24, 2014. The authors have no funding, financial relationships, or conflicts of interest to disclose. Send correspondence to Surjeet Pohar, MD, Indiana University Health East, 6845 Rama Drive, Indianapolis, IN 46219. E-mail: spohar@ iuhealth.org DOI: 10.1002/lary.25117
Laryngoscope 00: Month 2015
treatment. PEG placement has been shown to have minimal complications and to effectively limit weight loss.2 In a survey of head and neck cancer patients, Terrell et al. demonstrated that continued feeding tube use after treatment had the greatest negative effect on quality of life of any factor analyzed, including total laryngectomy and more than two comorbid conditions.3 We decided to use our patient data to determine whether PEG tube placement versus reliance on oral intake prior to CRT affected long-term severe dysphagia and weight loss during CRT. We also wanted to determine what the stricture rate was in our patients and whether there was a correlation between stricture development and long-term severe dysphagia.
MATERIALS AND METHODS The Eastern Virginia Medical School Cancer registry was queried to identify all patients with a new head and neck cancer diagnosis from January 1, 2001 to July 1, 2008. A total of 223 patients with stage III or IV squamous cell head and neck (SCCHN) carcinoma were identified. After exclusion of patients treated with primary surgical therapy, those receiving CRT at an outside facility, those requiring PEG tube feeding prior to therapy, those treated with radiation therapy (XRT) only, and those lost to follow-up, a total of 129 patients were included. All patients were reviewed for demographic information including age, primary site, stage, and Zubrod score at presentation. Patients were classified by Zubrod into those with Zubrod >1 (unable to work) or Zubrod of 0 or 1 (able to do at least light work). Ninety-nine patients had a prophylactic PEG tube placed (PPEG group), whereas 30 did not (initial oral intake group). Of this group, 22 died of disease within 1 year, two died of other causes, and one had no follow-up swallowing status reported.
Pohar et al.: PEG Tube Dependence Following Chemoradiation
1
TABLE I. Radiation Therapy Details in Patients with >1-Year Follow-up. Prophylactic, n 5 79
No Prophylactic PEG, n 5 25
TABLE III. Patient Characteristics for Those With >1-Year Follow-up (n 5 104). Prophylactic PEG,No. (%)
PO, No. (%)
Age, yr, median
57
53
Follow up (months), median
36
48
57 (72)
22 (28)
22 (88)
3 (12)
P
Technique Standard, n (%) IMRT, n (%)
69 (66) 10 (10)
20 (19) 5 (5)
Median radiation dose, cGy
7,030
7,030
Treatment delay, median, d
3
4
.35
Sex Male
IMRT 5 intensity-modulated radiation therapy; PEG 5 percutaneous endoscopic gastrostomy.
Thus, 104 patients were alive, had at least 12 months of followup, and had swallowing status recorded at the last follow-up. Swallowing status at the last follow-up was classified as PEG dependent or diet via oral intake. In this group of 104 patients, 79 had a PPEG and 25 began therapy with only oral intake. All patients underwent evaluation by a head and neck surgeon followed by a staging endoscopy. The need for gastrostomy tube placement was made on an individual basis based on tumor site, patient condition, and patient wishes. Sixty-three of 70 patients (90%) managed by one surgeon had a prophylactic PEG tube placed versus 36 of 59 (61%) managed by the other surgeon (P 5.002). All patients then underwent definitive treatment consisting of external beam radiation therapy with concurrent chemotherapy at Sentara Norfolk General Hospital or an affiliated facility. Patients were not given any specific instruction on swallowing exercises; however, most patients did see a dietitian who encouraged adequate nutritional intake either orally or by tube feeding. Patients in both groups were encouraged to continue oral intake. Most patients underwent standard two-dimensional or three-dimensional conformal radiation therapy. A smaller proportion of patients were treated with intensity-modulated radiation therapy (IMRT) (Table I). Various chemotherapy regimens were used, with weekly carboplatin/taxol being the most common (Table II). All patients were followed by radiation oncology, hematology-oncology, and otolaryngology for complaints of dysphagia, decreased oral intake, weight loss, and hospital admission. Those who did not have a prophylactic PEG (initial oral intake group) tube, had a therapeutic PEG tube placed if they had inadequate or no oral intake during treatment. The patients’ clinical courses were followed to assess the primary outcomes of swallowing/diet at 6 months, 1 year, and at last follow-up, as well as the need for esophageal dilation and treatment delay. Treatment delay was calculated by estimating expected date of completion of XRT treatment. Additional outcomes of local and regional recurrence as well as survival were
Female Race African American Caucasian
23 (85)
4 (15)
56 (73)
21 (27)
Overall stage Stage III
21 (27)
8 (32)
Stage IV
58 (73)
17(68)
T stage T0
1 (1)
3 (12)
T1
11 (14)
3 (12)
T2 T3
17 (22) 27 (34)
8 (32) 6 (24)
T4
23 (29)
5 (20)
N stage N0
16 (20)
6 (24)
N1
14 (18)
4 (16)
N2 N3
40 (51) 9 (11)
11 (44) 4 (16)
Primary site larynx/hypopharynx
24 (30)
3 (12)
Other Zubrod
55 (70)
22(88)
0–1
65 (86)
21 (88)
>1
11 (14)
3 (12)
P Value
.11
.19
.60
.09
.87
.07
.81
PEG 5 percutaneous endoscopic gastrostomy; PO 5 oral intake.
also collected. All aspects of the research were approved by the Eastern Virginia Medical School Institutional Review Board. SPSS statistical package version 18.0 was used for analysis (SPSS Inc., Chicago, IL).
RESULTS Patient Demographics
TABLE II. Chemotherapy in 104 Patients With >1-year Follow-up. Chemotherapy Regimen
Cisplatin weekly High-dose cisplatin every 3 weeks
No. of Patients
14 2
Other cisplatin-only regimen
2
Carboplatin/taxol weekly Cetuximab weekly
52 2
Carboplatin weekly
2
Other combination
27
Laryngoscope 00: Month 2015
2
Patient characteristics are described in Table III. Overall, out of 104 patients who had at least 1 year of follow-up, 79 patients underwent prophylactic PEG tube placement at the time of pan-endoscopy (PPEG group). An additional 25 patients received no feeding conduit at the time of staging (initial oral intake group). Of these 25 patients, nine required gastrostomy tube placement during CRT, and two patients had tubes placed after CRT (therapeutic PEG). No significant differences between the PPEG and initial oral intake groups with regard to age, sex, race, Zubrod score, T stage, N stage, and overall stage (Table III) were found. Median followup for these 104 patients was 39 months. Pohar et al.: PEG Tube Dependence Following Chemoradiation
TABLE IV. Five-Year Kaplan-Meier Adjusted Control and Survival. Prophylactic PEG
No Prophylactic PEG
P value
Local control
84.8%
80.0%
0.58
Regional control
88.9%
93.3%
0.73
Overall survival Disease specific survival
55.6% 65.5%
73.3% 78.6%
0.09 0.24
PEG 5 percutaneous endoscopic gastrostomy.
Hypopharyngeal or laryngeal carcinoma patients were more likely to receive a PPEG than other primary site patients (P 5.07).
Survival In comparing the survival for all 129 patients in the prophylactic PEG versus initial oral intake groups, no difference was found in terms of 5-year estimates of local control, regional control, overall survival, and disease-specific survival (Table IV).
Radiation Therapy No significant differences were observed in regard to radiation method, overall dose, or treatment delay for the 104 patients with at least 1-year follow-up (Table I).
Swallowing The surrogate endpoint used for poor swallowing ability was PEG tube dependence. There was no difference in PEG tube dependence between the PPEG and oral intake groups at 6 months (P 5.27). However, PEG tube dependence at last follow-up at least 1 year out
from therapy was 4% in those with no initial PEG versus 23% with prophylactic PEG (P 5.03). Other factors associated with PEG tube dependence in univariate analysis were poor functional status defined by a Zubrod score >1 and higher T stage (Table V). Other factors examined, including stage, N stage, and smoking status, were not significant (Table V). Statistically significant factors and subsite were included in a multivariate analysis of the 104 patients with more than 1-year follow-up. This demonstrated that PEG tube dependence was associated with higher T stage, Zubrod score >1, and use of prophylactic PEG feeding (Table V). Four patients had dysphagia prior to CRT, and all had a PPEG (one did not have a 1-year follow-up). Excluding the three with >1-year follow-up from the univariate and multivariate analysis did not change our findings (Table V). The raw data for multivariate analysis of the diet at last followup versus prognostic factor are shown in Table VI. The need for esophageal dilation is an important complication and comorbidity of CRT. Twenty-five of the 99 patients (25.3%) receiving a prophylactic gastrostomy tube required esophageal dilation compared to four patients (13.3%) in the initial oral intake group (P 5.22). There was no correlation between development of stricture requiring dilation and PEG dependence at last follow-up (n 5 129, P 5.65), but there was a trend toward a correlation when only including patients with >1-year follow-up (n 5 104, P 5.11). Data on weight loss during treatment for the entire group of 129 patients studied were available in 82 patients who had a prophylactic PEG tube and 26 patients who did not. There was no significant difference in the average weight decline of 8 kg (range, 13.8 kg to 227.7 kg) from 78 kg to 70 kg in patients with a prophylactic PEG tube and 8 kg from 90 kg to 82 kg (range, 12.2 to 217.1 kg) in those relying initially on oral intake (P 5.95).
TABLE V. Univariate and Multivariate Analysis of Prognostic Factors Versus PEG Dependence at Last Follow-up >1 Year (n 5 104), Excluding Patients With Pretreatment Dysphagia (n 5 101). Univariate, n 5 104, P Value
Multivariate, n 5 104, P Value
Univariate Excluding Dysphagia, n 5 101, P Value
Multivariate Excluding Dysphagia, n 5 101, P Value
Age Sex
.60 .40
— —
.57 .31
— —
Race (African American vs. white)
.24
—
.27
—
T stage N stage
.008 .29
.01 —
.009 .25
.017
Stage (III vs. IV)
.13
—
.09
—
Smoking Alcohol use
.60 .63
— —
.60 .63
— —
Variable
Initial PEG vs. no PEG Zubrod (>1 vs. 0–1) Laryngeal/hypopharyngeal primary
.03
.04
.03
.04
.000001 0.97
.000001 0.10
.000002 0.86
.000001 0.13
PEG 5 percutaneous endoscopic gastrostomy.
Laryngoscope 00: Month 2015
Pohar et al.: PEG Tube Dependence Following Chemoradiation
3
TABLE VI. Multivariate Analysis Raw Data for Swallowing at Last Follow-up >1 Year (n 5 104). Swallowing, No. PEG Dependent
PO
Initial PEG
18
61
No initial PEG Zubrod score
1
24
10
76
9
5
0
0
4
1 2
0 2
14 23
3
10
23
7
21
Laryngeal/hypopharyngeal
5
22
Other primary site
14
63
0–1 >1 T stage
4 Primary site
P-value
0.04
0.000001 0.01
0.10
PEG 5 percutaneous endoscopic gastrostomy; PO 5 oral intake.
DISCUSSION Weight Loss and Swallowing The need for prophylactic PEG placement in CRT remains a controversial topic. It is undeniable that CRT results in significant morbidity, with up to 86% of patients experiencing grade 3 or 4 mucositis.2 As a result, some authors have recommended universal placement of percutaneous gastrostomy tubes in all head and neck cancer patients prior to CRT.2 The primary argument for this approach is the prevention of malnutrition, weight loss, and subsequent hospitalization. Lee et al. examined the effectiveness of prophylactic PEG tube placement in 88 patients with advanced head and neck cancer.2 Of their cohort, 41% of patients received a gastrostomy tube prior to beginning CRT or twice-daily radiation alone. Those patients not receiving a gastrostomy tube lost an average of 7.0 kg versus 3.1 kg in the PPEG group (P 1 year from treatment in patients who initially had only oral intake. Several authors have tried to determine what treatment or patient attributes are associated with long-term dysphagia in CRT patients. Eisbruch et al. have found a correlation between radiation dose to the pharyngeal constrictors and long-term dysphagia.6 Caudell et al. and others have shown that laryngeal/hypopharyngeal cancer patients are more likely than other head and neck cancer patients to have long-term swallowing difficulties.7 They also found that a median dose >50 Gy to the larynx or inferior pharyngeal constrictor correlated with a higher risk of long-term severe dysphagia. When they excluded laryngeal/hypopharyngeal patients, they found that higher doses to the middle and superior constrictor also correlated with increased risk of developing long-term severe dysphagia.8 Sanguineti et al. found that higher oral cavity dose correlates with higher longterm dysphagia rates.9 Chen et al. found that 1-year PEG tube dependence was 21% in their PPEG patients versus zero in the initial oral intake group (P 1, and higher T stage to predict for PEG tube dependence at last follow-up >1 year. A summary of various series reporting on swallowing dysfunction, stricture, and weight loss is presented in Table VII. Our patients did not have routine consultation with a speech pathologist to initiate a program of prophylactic swallowing exercises to prevent dysphagia. Most studies of prophylactic swallowing exercises suggest an improvement in dysphagia; however, they generally report only 3 to 6 months follow-up and have small datasets.11,12 Hutcheson et al. recently reported on 497 patients treated with XRT alone or CRT for head and neck cancer.13 They found on multivariate analysis that maintaining swallowing exercises or a full diet throughout therapy each predicted for a higher likelihood of maintaining long-term regular dietary intake. Our current practice is to support patients with analgesics and counsel them to maintain oral intake even if they have a PEG tube, and to refer to speech pathology at the time of diagnosis for formal swallow evaluation and initiation of prophylactic swallowing exercises during treatment. The findings of the randomized trial by Silander et al. of a lower rate of long-term PEG tube dependence in the PPEG group is the opposite of those in our study and in Chen et al.’s data.4 Further studies are needed to see if our findings are confirmed in patients treated with CRT using IMRT only. We found that Zubrod score >1 (unable to work) correlated with long-term PEG tube dependence. We are not sure why this is, but perhaps patients with more functional disability also have less ability to overcome tissue injury from CRT. Higher T-stage patients may have had higher rates of long-term swallowing dysfunction because they had a larger volume of tissue treated to a high dose, resulting in more tissue being damaged beyond recovery (Table VI).
Stricture We found a stricture rate requiring dilation of 25% in our PPEG group, similar to the 30% with PPEG reported by Chen et al.10 Clavel et al. and Mekhail et al., using a policy of initial oral intake with a therapeutic NG tube as needed, found the rate of stricture to be only 3% and 4%, respectively.5,14 They hypothesized that the NG tube served a stenting function. However, Chen et al. found a 6% stricture rate, and we found a Laryngoscope 00: Month 2015
13% rate for patients on initial oral intake during CRT with therapeutic PEG placement as needed.10 Furthermore, NG tubes are uncomfortable for patients to have for the median of 40 days in the Clavel study and 8 weeks in Mekhail et al.’s group.5,14 Mekhail et al. found that 23% of therapeutic PEG tube patients required pharyngoesophageal dilation compared with only 4% of patients treated with a therapeutic NG feeding tube during CRT.14 The authors did state that patients selected for PEG tube placement were more malnourished than those treated with NG tubes. A summary of complications in the various series is presented in Table VII.
Treatment Delay The effect of treatment delay on overall survival was examined by Fesinmeyer et al. by reviewing Medicare data from 3,864 head and neck cancer patients.15 Their research demonstrated that in laryngeal cancer, a treatment delay of 5 to 30 days resulted in a 68% increased risk of death compared with patients with no interruptions. Some have thought prophylactic PEG would reduce treatment delays and thus improve effectiveness of CRT.2 Our results, similar to those of Chen et al.,10 failed to find a significant difference in treatment delay or locoregional control between the PPEG group and oral intake group. Considering these findings, it is questionable whether universal PEG tube placement is necessary or beneficial in regard to treatment delays during chemoradiation. When evaluating the need for prophylactic PEG tube placement, one must consider that placement is not entirely benign. Although PEG tube insertion presents a small risk of complication, the risk is real. Complications such as colon perforation can occur. In addition to immediate comorbid effects, some researchers have demonstrated increased esophageal toxicity in those patients undergoing PEG tube placement.10,14 It is important to acknowledge that as with most research on this topic, our collection method was retrospective with obvious limitations. These include imbalance between number of patients having a PPEG and initial oral intake, and selection bias so that a larger proportion of high–T-stage patients and hypopharyngeal/ laryngeal patients received PPEG (Table III). Ideally, this would be corrected by completing a randomized trial stratifying patients for known prognostic factors from our study and other retrospective studies. Also, such a study would use current radiation techniques of IMRT and image-guided radiation therapy. However, the only randomized trial comparing prophylactic PEG tube versus oral intake in exclusively CRT patients was closed because of low patient accrual.16
CONCLUSION Patients with head and neck carcinoma treated with CRT were found to have higher rates of long-term PEG tube dependence if they had a PPEG, had a higher T stage, or had a Zubrod score >1. Prophylactic PEG tube placement did not reduce treatment breaks during the radiation course, affect overall survival, or reduce weight Pohar et al.: PEG Tube Dependence Following Chemoradiation
5
loss during CRT. Stricture rates were lower in patients started initially on oral intake than with prophylactic PEG. It seems that initial oral intake with therapeutic PEG as needed is the better approach. Clearly, further prospective research is needed to develop standards and protocols for the placement of prophylactic PEG placement.
BIBLIOGRAPHY 1. Kulbersh BD, Rosenthal EL, McGrew BM, et al. Pretreatment, preoperative swallowing exercises may improve dysphagia quality of life. Laryngoscope 2006;116:883–886. 2. Lee JH, Machtay M, Unger LD, et al. Prophylactic gastrostomy tubes in patients undergoing intensive irradiation for cancer of the head and neck. Arch Otolaryngol Head Neck Surg 1998;124:871–875. 3. Terrell JE, Ronis DL, Fowler KE, et al. Clinical predictors of quality of life in patients with head and neck cancer. Arch Otolaryngol Head Neck Surg 2004;130:401–408. 4. Silander E, Nyman J, Bove M, Johansson L, Larsson S, Hammerlid E. Impact of prophylactic percutaneous endoscopic gastrostomy on malnutrition and quality of life in patients with head and neck cancer: a randomized study. Head Neck 2012;34:1–9. 5. Clavel S, Fortin B, Despres P, et al. Enteral feeding during chemoradiotherapy for advanced head-and-neck cancer: a single-institution experience using a reactive approach. Int J Radiat Oncol Biol Phys 2011;79:763–769. 6. Eisbruch A, Kim HM, Feng FY, et al. Chemo-IMRT of oropharyngeal cancer aiming to reduce dysphagia: swallowing organs late complication probabilities and dosimetric correlates. Int J Radiat Oncol Biol Phys 2011;81:e93–e99. 7. Caudell JJ, Schaner PE, Meredith RF, et al. Factors associated with longterm dysphagia after definitive radiotherapy for locally advanced headand-neck cancer. Int J Radiat Oncol Biol Phys 2009;73:410–415.
Laryngoscope 00: Month 2015
6
8. Caudell JJ, Schaner PE, Desmond RA, Meredith RF, Spencer SA, Bonner JA. Dosimetric factors associated with long-term dysphagia after definitive radiotherapy for squamous cell carcinoma of the head and neck. Int J Radiat Oncol Biol Phys 2010;76:403–409. 9. Sanguineti G, Gunn GB, Parker BC, Endres EJ, Zeng J, Fiorino C. Weekly dose-volume parameters of mucosa and constrictor muscles predict the use of percutaneous endoscopic gastrostomy during exclusive intensitymodulated radiotherapy for oropharyngeal cancer. Int J Radiat Oncol Biol Phys 2011;79:52–59. 10. Chen AM, Li BQ, Lau DH, et al. Evaluating the role of prophylactic gastrostomy tube placement prior to definitive chemoradiotherapy for head and neck cancer. Int J Radiat Oncol Biol Phys 2010;78:1026–1032. 11. Carnaby-Mann G, Crary MA, Schmalfuss I, Amdur R. "Pharyngocise": randomized controlled trial of preventative exercises to maintain muscle structure and swallowing function during head-and-neck chemoradiotherapy. Int J Radiat Oncol Biol Phys 2012;83:210–219. 12. van der Molen L, van Rossum MA, Burkhead LM, Smeele LE, Rasch CR, Hilgers FJ. A randomized preventive rehabilitation trial in advanced head and neck cancer patients treated with chemoradiotherapy: feasibility, compliance, and short-term effects. Dysphagia 2011;26:155–170. 13. Hutcheson KA, Bhayani MK, Beadle BM, et al. Eat and exercise during radiotherapy or chemoradiotherapy for pharyngeal cancers: use it or lose it. JAMA Otolaryngol Head Neck Surg 2013;139:1127–1134. 14. Mekhail TM, Adelstein DJ, Rybicki LA, Larto MA, Saxton JP, Lavertu P. Enteral nutrition during the treatment of head and neck carcinoma: is a percutaneous endoscopic gastrostomy tube preferable to a nasogastric tube? Cancer 2001;91:1785–1790. 15. Fesinmeyer MD, Mehta V, Blough D, Tock L, Ramsey SD. Effect of radiotherapy interruptions on survival in medicare enrollees with local and regional head-and-neck cancer. Int J Radiat Oncol Biol Phys 2010;78: 675–681. 16. Corry J, Poon W, McPhee N, et al. Prospective study of percutaneous endoscopic gastrostomy tubes versus nasogastric tubes for enteral feeding in patients with head and neck cancer undergoing (chemo)radiation. Head Neck 2009;31:867–876.
Pohar et al.: PEG Tube Dependence Following Chemoradiation
