Swallowing Outcomes Following Laryngectomy and Pharyngolaryngectomy

Objectives  To determine the incidence of dysphagia (defined as the inability to manage a diet of normal consistencies) at hospital discharge and beyond 1 year postsurgery and examine the impact of persistent dysphagia on levels of disability, handicap, and well-being in patients.

Design  Retrospective review and patient contact.

Setting  Adult acute care tertiary hospital.

Patients  The study group, consecutively sampled from January 1993 to December 1997, comprised 55 patients who underwent total laryngectomy and 37 patients who underwent pharyngolaryngectomy with free jejunal reconstruction. Follow-up with 36 of 55 laryngectomy and 14 of 37 pharyngolaryngectomy patients was conducted 1 to 6 years postsurgery.

Main Outcome Measures  Number of days until the resumption of oral intake; swallowing complications prior to and following discharge; types of diets managed at discharge and follow-up; and ratings of disability, handicap, and distress levels related to swallowing.

Results  Fifty four (98%) of the laryngectomy and 37 (100%) of the pharyngolaryngectomy patients experienced dysphagia at discharge. By approximately 3 years postsurgery, 21 (58%) of the laryngectomy and 7 (50%) of the pharyngolaryngectomy patients managed a normal diet. Pharyngolaryngectomy patients experienced increased duration of nasogastric feeding, time to resume oral intake, and incidence of early complications affecting swallowing. Patients experiencing long-term dysphagia identified significantly increased levels of disability, handicap, and distress. Patients without dysphagia also experienced slight levels of handicap and distress resulting from taste changes and increased durations required to complete meals of normal consistency.

Conclusions  The true incidence of patients experiencing a compromise in swallowing following surgery has been underestimated. The significant impact of impaired swallowing on a patient's level of perceived disability, handicap, and distress highlights the importance of providing optimal management of this negative consequence of surgery to maximize the patient's quality of life.

DYSPHAGIA, or impaired swallowing function, has been established as a predominant negative sequelae following laryngectomy and pharyngolaryngectomy surgery.^1-5 Despite this, few investigations have specifically examined the nature of the presenting dysphagia, including postoperative and long-term complications, patterns of recovery, or long-term dietary outcomes. In addition, the documented incidence figures for dysphagia following total laryngectomy and pharyngolaryngectomy are noted to vary widely from study to study. Presently, the incidence of dysphagia following total laryngectomy has been reported to range from 10%⁶ to 60%.⁷ Similarly, the overall reported incidence of dysphagia following pharyngolaryngectomy with free jejunal graft reconstruction currently ranges from 2%⁸ to 58%.⁹ The inconsistency in these figures in both patient populations seems to stem from the use of various definitions of "successful" swallowing. Many authors have categorized successful swallowing as the ability to achieve oral intake regardless of consistency, eg, "successful swallowing is the ability to maintain nutrition without tube feeding."¹⁰(p396) Others have included the ability to swallow modified consistencies as their optimal outcome, eg, "adequate swallowing is the ability to tolerate a soft diet."¹¹(p954) Although such broad definitions of swallowing success may be adequate when evaluating the outcomes of different surgical techniques, they fail to accurately reflect the true number of patients postsurgery who are unable to resume their normal, premorbid swallowing function. Consequently, it is possible that the figures reported in the literature may have underestimated the number of patients who may experience difficulties and associated distress related to altered swallowing function postsurgery.

The psychosocial aspects of dysphagia, which arise as a consequence of long-term swallowing dysfunction, and their ultimate impact on a patient's quality of life have received minimal attention in the literature. The restoration of the ability to swallow and eat normally by mouth is critical for full social rehabilitation of patients,¹² and the inability for some patients to return to a normal diet following surgery can have a negative impact on quality of life.² Ackerstaff et al¹ evaluated the functional disorders and lifestyle changes following total laryngectomy and found that as many as 25% of patients report alterations to their diet, including avoidance of certain consistencies as well as modifications to their style of eating. Investigations have also shown that these monotonous dietary changes can often lead to reduced appetite and weight loss, which ultimately result in poor quality of life.²

It is, therefore, the aim of the present study to document the incidence and severity of dysphagia both in the immediate postsurgical and long-term postsurgical phases of patients following either a total laryngectomy or pharyngolaryngectomy procedure. In this study, dysphagia or a swallowing impairment has been defined as any inability to manage a full diet of normal consistencies. In light of the limitations of the definitions of dysphagia used in existing research, we hypothesize that the true incidence of dysphagia in the total laryngectomy and pharyngolaryngectomy population has been underestimated. Our study also aims to specifically evaluate the impact of long-term swallowing dysfunction on levels of perceived handicap and distress in the total laryngectomy and pharyngolaryngectomy patient populations. It is hypothesized that any patient who has not resumed the ability to manage a normal diet postsurgery will identify themselves as having increased levels of disability, handicap, and distress compared with those managing a normal diet.

Subjects included all patients admitted to the Royal Brisbane Hospital, Brisbane, Australia, between January 1993 and December 1997 for either a total laryngectomy or pharyngolaryngectomy with free jejunal interposition reconstruction. Patients were excluded from the present study if they had undergone oral surgery or had a coexisting neurological impairment or injury that may affect swallowing. We identified 55 laryngectomy and 37 pharyngolaryngectomy patients who were eligible. Of the laryngectomy patients, 50 were men and 5, women (mean ± SD age at surgery, 63 years ± 8 years 10 months; range, 39-82 years). Of the pharyngolaryngectomy patients, 34 were men and 3, women (mean ± SD age at surgery, 59 years 9 months ± 8 years 9 months; range, 39-77 years). In the laryngectomy group, 17 (31%) had received prior radiotherapy; 27 (49%), postoperative radiotherapy; 2 (4%), both; and 9 (16%), no radiotherapy. Of the pharyngolaryngectomy patients, 5 (14%) had received prior radiotherapy; 29 (78%), postoperative radiotherapy; 1 (3%), both; and 2 (5%), no radiotherapy.

Ethical clearance for data collection was obtained from the Royal Brisbane Hospital and The University of Queensland. The medical records and speech pathology files of all patients were reviewed, and details pertaining to medical and surgical characteristics, postoperative swallowing complications, swallowing assessments and management, and postsurgical dietary outcomes were collated. Swallowing complications were defined as any symptom observed on either clinical swallowing assessments or radiological investigations that resulted in a negative effect on the patient's swallowing ability. Complications were divided into those that occurred within the first month postsurgery (referred to as early postsurgical complications) and those that occurred beyond the first month postsurgery (referred to as late postsurgical complications). The immediate but transient effects of radiotherapy were not included as a complication in this study because those data were inconsistently reported in the medical files. Six categories of dietary consistencies were used to classify the dietary status of each patient: normal, soft selective (soft options of a normal diet), soft mechanical (soft chewable consistencies), soft puree (vitamized foods [foods blended with additional gravy or sauce]), liquid puree (liquefied foods), and nonoral feeding. For the purposes of this study, clinically significant dysphagia was defined as the inability to tolerate a normal diet, ie, able to swallow all liquid and solid foods without any texture alteration required and no requirement for any supplementary nonoral nutrition.

On completion of the retrospective medical record review phase of the investigation, all patients were contacted to discuss their long-term swallowing outcomes. At that time, all patients were at a minimum of 1 year and a maximum of 6 years postsurgery. Only 36 of the 55 laryngectomy (30 men and 6 women; mean ± SD years postsurgery, 3 years 3 months ± 1 year 6 months) and 14 of the 37 pharyngolaryngectomy patients (13 men and 1 woman; mean ± SD years postsurgery, 3 years 2 months ± 1 year 2 months) could be contacted. Of the remainder, 32 of the 92 patients had deceased, and 10 of 92 had invalid contact details. During the interview, patients identified (1) current dietary status and (2) levels of perceived swallowing disability, handicap, and well-being/distress using the Therapy Outcome Measure Dysphagia Scale (TOM),¹³ a series of 6-point scales for which 0 indicates a highly negative result (extreme difficulties) and 5 indicates a highly positive result (absence of difficulties).

The mean ± SD period of hospitalization was 19.0 ± 18.78 days (range, 5-127 days) in the laryngectomy group and 23.5 ± 15.10 days (range, 11-78 days) in the pharyngolaryngectomy group. Statistical comparison revealed no significant difference between the duration of hospitalization of the 2 groups. A number of patients experienced multiple swallowing-related complications within the first month postsurgery, with 19 complications identified in 15 (27%) of the 55 laryngectomy patients vs 35 in 24 (65%) of the 37 pharyngolaryngectomy patients (Table 1). Statistical analysis revealed a significantly (z = 3.364; P<.001) higher incidence of early swallowing complications in the pharyngolaryngectomy group. There was no significant difference between the number of late complications reported for 20 (36%) of the 55 laryngectomy patients and 15 (40%) of the 37 pharyngolaryngectomy patients (Table 1).

Postoperatively, all patients received initial nutrition via nasogastric feeding to allow for the recovery of surgical tissue. At a mean ± SD of 11.7 ± 7.31 days (range, 6-75 days) following surgery, each patient was clinically assessed using either a modified barium swallow (7 patients [8%]), diatrizoate meglumine (Gastrografin; Bristol-Meyers Squibb Co, Princeton, NJ) swallow (37 patients [40%]), or blue dye swallow (48 patients [52%]) to assess the feasibility of initiating oral intake. For 49 (89%) of the 55 laryngectomy patients and 29 (78%) of the 37 pharyngolaryngectomy patients who presented with no complications that would inhibit the resumption of oral intake, the mean ± SD duration to oral alimentation was 10.7 ± 1.98 days in the laryngectomy group and 12.1 ± 2.12 days in the pharyngolaryngectomy group. Statistical comparison revealed the duration to oral alimentation was significantly (t test = −2.886; P<.01) shorter in the laryngectomy group. The remaining 6 (11%) of the laryngectomy and 8 (22%) of the pharyngolaryngectomy patients were identified with fistulae or wound breakdown on the immediate radiographic evaluation. The 6 laryngectomy patients (11%) received extended nasogastric feeding for a mean ± SD of 24.7 ± 14.47 days. Five of the 8 pharyngolaryngectomy patients received extended nasogastric feeding for a mean ± SD of 27.2 ± 9.63 days, while the remaining 3 pharyngolaryngectomy patients (8%) received percutaneous endoscopic gastronomy feeding for a mean ± SD of 65.3 ± 23.44 days.

Following discharge, a further 15 (27%) of the laryngectomy and 6 (16%) of the pharyngolaryngectomy patients who were identified with late complications required periods of nonoral nutrition. Twelve (22%) of the laryngectomy and 2 (5%) of the pharyngolaryngectomy patients received nasogastric feeding for a mean ± SD period of 36.6 ± 40.09 days and 17.0 ± 12.73 days, respectively. One pharyngolaryngectomy patient (3%) with long-term complications required additional protein and energy supplements indefinitely, and the remaining 3 (5%) laryngectomy and 3 (8%) pharyngolaryngectomy patients required feeding via percutaneous endoscopic gastronomy for a mean ± SD period of 23.3 ± 17.93 days and 16.3 ± 13.05 days, respectively.

At hospital discharge, only 1 laryngectomy patient (2%) was tolerating a normal diet. According to our definition of dysphagia, the remaining 54 laryngectomy patients (98%) and 37 pharyngolaryngectomy patients (100%) were classified as dysphagic owing to their inability to manage foods of normal consistency (Table 2). Statistical analysis revealed no significant difference between the incidence of dysphagia identified in the 2 groups at discharge.

The long-term follow-up was conducted at a mean ± SD of 3 years 3 months ± 1.50 years postsurgery for laryngectomy patients and 3 years 2 months ± 1.17 years for pharyngolaryngectomy patients. At that time, 21 (58%) of the 36 laryngectomy and 7 (50%) of the 14 pharyngolaryngectomy patients with dysphagia continued to be classified as having dysphagia (Table 3). Statistical comparison of proportions revealed no significant difference between the incidence of long-term dysphagia between the 2 groups.

At follow-up, there was no significant difference (P = .39) between the mean ± SD TOM scores for levels of swallowing disability experienced by either the total laryngectomy (4.39 ± 0.84) or pharyngolaryngectomy (4.21 ± 0.97) group. Similarly, the mean ± SD TOM scores for perceived levels of handicap in the total laryngectomy (4.14 ± 0.96) and pharyngolaryngectomy (3.93 ± 0.73) patients and the levels of well-being/distress in the total laryngectomy (4.25 ± 1.00) and pharyngolaryngectomy (4.36 ± 0.63) groups were not significantly different (P = .23 and .89, respectively).

Within each surgical group, statistical comparisons were conducted between the patients with and without dysphagia (Table 4). In the laryngectomy group, results revealed significantly (P<.001) higher levels of disability, handicap, and distress in the 15 patients with dysphagia compared with the 21 patients without dysphagia. In the pharyngolaryngectomy patient group, there were insufficient group numbers (7 with dysphagia and 7 without dysphagic) to validate statistical comparison.¹⁴ However, an examination of the group mean TOM scores revealed a consistently higher level of perceived disability, handicap, and distress levels in the pharyngolaryngectomy patients with dysphagia. Interestingly, although the patients without dysphagia in both groups rated themselves as having no swallowing disability, some patients perceived that they experienced a slight level of swallowing handicap and associated distress. These ratings were most frequently attributed by the patient to an increased duration necessary to eat a meal of normal consistency and disruptions to taste and smell postoperatively.

Despite the differential nature of the 2 surgical procedures, both the laryngectomy and pharyngolaryngectomy patient groups presented with comparable levels of dysphagia at the point of discharge postsurgery. Specifically, the present study revealed that 54 (98%) of the laryngectomy and 37 (100%) of the pharyngolaryngectomy patients presented with persistent swallowing impairments that prevented the management of normal dietary consistencies at discharge. As a consequence of using more stringent criteria to define swallowing impairment, the present percentages are significantly higher than previous reports of postoperative dysphagia for either the laryngectomy or pharyngolaryngectomy population.^7,9 The definitions used in the present study, however, more accurately reflect the true existence or absence of swallowing compromise. In light of this, the current figures would seem to represent a more accurate indication of the true incidence of patients who experience swallowing impairments, regardless of severity, following total laryngectomy and pharyngolaryngectomy.

Following a mean period of 3 years postdischarge, longitudinal evaluation revealed that over half of the patients in both surgical groups had achieved optimal swallowing outcomes and resolution of the ability to manage a normal diet. This represented a dramatic increase from 2% to 58% in the laryngectomy group and 0% to 50% in the pharyngolaryngectomy group. In addition, 49 (89%) of the laryngectomy and 29 (78%) of the pharyngolaryngectomy patients were managing a soft diet or better at follow-up, and no patient continued to require nonoral feeding support. These findings demonstrate good long-term functional results for both surgical procedures consistent with previous literature.^8,15,16 The present incidence figures, however, remain on the conservative side of the few existing reports of long-term swallowing outcomes.^8,16 Hillman et al¹⁶ reported that 76% of their laryngectomy patients at 24 months postsurgery managed a normal diet, while Julieron et al⁸ similarly found that 76% of pharyngolaryngectomy patients could manage foods of normal consistency at 12 months postsurgery.

Although the incidence of dysphagia at discharge and in the long-term did not differ between the 2 surgical groups, significant differences were noted in the early rehabilitation of oral intake between the groups. Specifically, the more extensive surgical procedure associated with pharyngolaryngectomy resulted in a significantly longer period of immediate postoperative nonoral nutrition and delays to the commencement of oral feeding compared with the laryngectomy group. The present study also revealed that the pharyngolaryngectomy patients experienced a significantly higher incidence of early complications related to swallowing dysfunction than the laryngectomy patients during the early postsurgical phase. The prominent complications observed in the group included nasal regurgitation and fistula formation, which have been reported in the literature as frequently occurring following pharyngolaryngectomy surgery with jejunal graft reconstruction.^17,18 Beyond the early postsurgical phase, however, the incidence of complications affecting swallowing function in the long-term did not differ between the 2 surgical groups, with less than 40% of patients experiencing some complication that developed beyond 1 month postsurgery.

The swallowing impairments noted in both patient groups impacted negatively on patients' quality of life. In particular, laryngectomy patients who presented with persistent dysphagia reported significantly higher levels of disability than those patients who could tolerate a normal diet. Those patients with a perceived disability reported an inability to manage normal dietary consistencies, the need to implement compensatory strategies to facilitate their swallowing, or the dependence on additional nutritional supplements. For laryngectomy patients with persistent swallowing impairments, the results also reflected poor quality of life with the perception of higher levels of handicap and distress than the laryngectomy patients who experienced normal long-term swallowing function. Laryngectomy patients with dysphagia reported that the prolongation of swallowing impairments following the postoperative phase had affected their ability to fulfill certain social, educational, and/or family roles; decreased their self-confidence and self-esteem; and resulted in poor ability to achieve their potential in certain situations. Considering that eating is a crucial component not only in sustaining nutritional needs, but also in socializing and enjoying lifelong traditions, all of these patients experienced impaired quality of life as a result of prolonged swallowing dysfunction.⁵ Furthermore, reduced quality of life was also evident in the increased reports of anger, frustration, embarrassment, concern, and withdrawal in laryngectomy patients with dysphagia, which was consistent with some reports that patients with cancer find eating to be more distressing than nourishing.¹⁹

At present, there is a sparsity of literature detailing the functional recovery of swallowing following pharyngolaryngectomy procedures with free jejunal graft reconstruction in the long-term or the impact of persistent dysphagic impairments on quality of life. This may be the result of a number of factors, including poor survival rates, which have been reported to fall as low as 15% to 35% following a 5-year period.^20,21 Poor longevity is further exacerbated by patient dropout from longitudinal studies, which can be expected following head and neck surgery.²² These 2 factors were demonstrated to affect the present study because insufficient patient numbers were available for follow-up in the pharyngolaryngectomy patient group to allow a definitive statistical analysis of the impact of long-term swallowing dysfunction on quality of life to be reported. Evaluation of the descriptive statistics revealed pharyngolaryngectomy patients who presented with long-term swallowing dysfunction, however, reported higher levels of perceived disability, handicap, and distress on average. This finding suggests that pharyngolaryngectomy patients with persistent dysphagia also experienced impaired social functioning and emotional repercussions as a result of poor swallowing ability.

Interestingly, in the present data some patients who presented with the ability to manage a normal diet at follow-up and reported no swallowing disability perceived that they continued to experience a mild level of handicap and distress related to their swallowing function. The interviews with these patients revealed that although the patients could manage a normal diet, compensations such as taking additional time to complete a meal or drinking increased amounts of liquids were viewed as negative changes. Other psychological and emotional issues (eg, the lack of "taste" of foods) was also reported by some as still impacting negatively on the social and pleasurable aspects of swallowing and eating postsurgery. These findings demonstrate that even mild alterations to normal eating and swallowing behaviors can alter a patient's perceptions about their quality of life. They also highlight how levels of disability do not always correspond to the perceived levels of handicap and distress.

Through the use of more stringent criteria that classified dysphagia as any inability to resume normal dietary status, the present study demonstrated a higher incidence of dysphagia at discharge and long-term follow-up in both the laryngectomy and pharyngolaryngectomy groups than was previously reported. The present study identified characteristic differences between the total laryngectomy and pharyngolaryngectomy patients with respect to the higher incidence of early postoperative complications and the extended duration to initial postsurgical dietary intake recorded in the pharyngolaryngectomy group. The incidence of dysphagia at discharge and long-term follow-up, however, was comparable for both groups.

The persistence of long-term swallowing impairments was noted to negatively affect a patient's quality of life, with more significant levels of disability, handicap, and distress reported by patients with dysphagia in both surgical groups. In addition, although the laryngectomy and pharyngolaryngectomy patients without dysphagia had resumed premorbid dietary status, some still perceived that they experienced increased levels of handicap and distress due to factors such as increased time required to complete a meal of normal consistency. This finding, and the results of the disability, handicap, and distress measures of the patients with dysphagia highlight that any degree of alteration to normal swallowing function can have a negative impact on a patient's quality of life. Consequently, it is important not to underestimate the postsurgical significance of any degree of swallowing impairment and its potential effect on a patient's level of well-being.

Accepted for publication August 29, 2001.

We would like to acknowledge the assistance of Melissa Timm, BspPath, Lynda Waymouth, BspPath, and Megan Bell, BspThy, for their assistance with the data collation and Barbara McCosker for her assistance with the medical records.

Corresponding author: Elizabeth Ward, PhD, Department of Speech Pathology and Audiology, The University of Queensland, St Lucia 4072, Australia (e-mail: Liz.Ward@mailbox.uq.edu.au).