This cohort study examines factors associated with 30-day postoperative complications, 90-day mortality, and 90-day functional decline and creates a preoperative risk stratification system for elderly patients undergoing head and neck ablation and reconstruction.
Key Points
Question
What characteristics are associated with worse outcomes for patients 80 years or older undergoing major head and neck ablative and reconstructive surgery?
Findings
In this multi-institutional cohort study of 376 patients from 17 academic centers, variables associated with worse outcomes were being 85 years or older, moderate or severe comorbidities, body mass index of less than 25, high frailty, duration of surgery, flap failure, additional operations, and surgery of the maxilla, oral cavity, or oropharynx. A novel risk stratification system incorporating preoperative patient factors is presented.
Meaning
Preoperative patient factors can be used to provide risk stratification information during preoperative counseling and treatment planning; the type of flap is not associated with worse outcomes in this population.
Abstract
Importance
Data regarding outcomes after major head and neck ablation and reconstruction in the growing geriatric population (specifically ≥80 years of age) are limited. Such information would be extremely valuable in preoperative discussions with elderly patients about their surgical risks and expected functional outcomes.
Objectives
To identify patient and surgical factors associated with 30-day postoperative complications, 90-day mortality, and 90-day functional decline; to explore whether an association exists between the type of reconstructive procedure and outcome; and to create a preoperative risk stratification system for these outcomes.
Design, Setting, and Participants
This retrospective, multi-institutional cohort study included patients 80 years or older undergoing pedicle or free-flap reconstruction after an ablative head and neck surgery from January 1, 2015, to December 31, 2017, at 17 academic centers. Data were analyzed from February 1 through April 20, 2019.
Main Outcomes and Measures
Thirty-day serious complication rate, 90-day mortality, and 90-day decline in functional status. Preoperative comorbidity and frailty were assessed using the American Society of Anesthesiologists classification, Adult Comorbidity Evaluation–27 score, and Modified Frailty Index. Multivariable clustered logistic regressions were performed. Conjunctive consolidation was used to create a risk stratification system.
Results
Among 376 patients included in the analysis (253 [67.3%] men), 281 (74.7%) underwent free-flap reconstruction. The median age was 83 years (range, 80-98 years). A total of 193 patients (51.3%) had 30-day serious complications, 30 (8.0%) died within 90 days, and 36 of those not dependent at baseline declined to dependent status (11.0%). Type of flap (free vs pedicle, bone vs no bone) was not associated with these outcomes. Variables associated with worse outcomes were age of at least 85 years (odds ratio [OR] for 90-day mortality, 1.19 [95% CI 1.14-1.26]), moderate or severe comorbidities (OR for 30-day complications, 1.80 [95% CI, 1.34-2.41]; OR for 90-day mortality, 3.33 [95% CI, 1.29-8.60]), body mass index (BMI) of less than 25 (OR for 30-day complications, 0.95 [95% CI, 0.91-0.99]), high frailty (OR for 30-day complications, 1.72 [95% CI, 1.10-2.67]), duration of surgery (OR for 90-day functional decline, 2.94 [95% CI, 1.81-4.79]), flap failure (OR for 90-day mortality, 3.56 [95% CI, 1.47-8.62]), additional operations (OR for 30-day complications, 5.40 [95% CI, 3.09-9.43]; OR for 90-day functional decline, 2.94 [95% CI, 1.81-4.79]), and surgery of the maxilla, oral cavity, or oropharynx (OR for 90-day functional decline, 2.51 [95% CI, 1.30-4.85]). Age, BMI, comorbidity, and frailty were consolidated into a novel 3-tier risk classification system.
Conclusions and Relevance
Important demographic, clinical, and surgical characteristics were found to be associated with postoperative complications, mortality, and functional decline in patients 80 years or older undergoing major head and neck surgery. Free flap and bony reconstruction were not independently associated with worse outcomes. A novel risk stratification system is presented.
Introduction
Improvements in health care and reduction in cigarette smoking have resulted in a significant increase in the life expectancy in the United States during the past several decades.1 Adults 85 years and older (termed the oldest old) are the fastest growing age group in the United States, with their numbers expected to triple from 2016 to 2060.2 The incidence of newly diagnosed head and neck squamous cell carcinoma in the elderly is also expected to increase by more than 60% by the year 2030.3 Cancer is the second most common cause of death in the elderly and similar to other cancer sites, head and neck cancer research often excludes the elderly due to disqualifying comorbidities.4
Elderly patients with head and neck cancer are often considered poor candidates for multimodality therapy owing to impaired functional status and preexisting comorbidities. Thus, elderly patients are less likely to receive standard of care treatment compared with their younger counterparts. For many locoregionally advanced head and neck cancers that tend to affect elderly patients (eg, cutaneous malignant neoplasms, oral cavity cancers, and parotid malignant neoplasms), extensive surgery and reconstruction constitute the only curative option, and palliative treatment options are lacking.
Although some data show that free tissue transfer can be safely performed in elderly patients,5,6,7,8 very little data in the literature are available to guide surgeons in the very old population (aged ≥80 years). It is challenging to estimate which elderly patients will tolerate extensive surgical procedures, what reconstructive procedures are best tolerated in this population (eg, free tissue transfer vs pedicle flap and bony vs soft tissue reconstruction), and what are the anticipated postoperative morbidity, mortality, and functional outcomes in this population. Such information would be invaluable when counseling patients and family and critical to preoperative surgical planning.
To answer this question, we performed a multi-institutional cohort study of patients 80 years or older who underwent a major head and neck surgical procedure and who received free tissue transfer or pedicle flap reconstruction during a 3-year period. The objectives of this study were (1) to identify patient and surgical factors associated with 30-day postoperative complications, 90-day mortality, and 90-day functional decline; (2) to explore whether the type of reconstructive surgery (eg, pedicle vs free flap and bony reconstruction vs soft tissue only) is associated with the outcomes; and (3) to create a preoperative risk stratification system for these outcomes that can then be used during preoperative patient counseling.
Methods
Study Design
A retrospective cohort review was performed from 17 North American academic centers enrolled through the American Head and Neck Society Reconstruction Committee. Institutional review board approval was obtained from all participating centers, which waived the need for informed consent for deidentified data. This study followed the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guideline.
All consecutive patients 80 years or older who underwent free tissue transfer or pedicle flap (defined as pectoralis, supraclavicular artery island, submental, pedicle latissimus, or trapezius flap) reconstruction from January 1, 2015, through December 31, 2017, were included. Preoperative demographics, comorbidity, frailty, baseline functional status, intraoperative surgical details, postoperative complications, flap loss, 90-day mortality, and 90-day functional status were obtained. The Modified Frailty Index (MFI)9 was used to assess frailty and the Adult Comorbidity Evaluation–27 (ACE-27) scoring system was used to assess comorbidities.10 American Society of Anesthesiologists classification was also obtained. Weight loss was defined as losing at least 4.5 kg in the preoperative period. The MFI consists of 11 dichotomous variables that are each equal to 1 point. The MFI score is calculated by dividing the sum of the variables by 11. Based on previous studies,11,12,13 an MFI cutoff of at least 0.25 (ie, presence of ≥3 frailty variables) was considered high frailty for the analysis. The ablative sites were categorized as (1) the scalp, facial skin, or parotid; (2) the maxilla, oropharynx, or oral cavity; and (3) the larynx, hypopharynx, or esophagus.
Study outcomes were 30-day serious complications (defined as medical complication, transfer to higher level of care, or 30-day readmission), 90-day mortality, and 90-day functional decline. Functional status was coded as independent, assisted living, or dependent/nursing home. Functional decline was defined as any patient starting at independent or assisted living status and declining to dependent status at 90 days postoperatively. Patients who were dependent at baseline or who died within 90 days were excluded from the functional decline analysis. Additional surgical procedures included any return to the operating room related to the original head and neck surgery during the index hospitalization.
Statistical Analysis
Data were analyzed from February 1 through April 20, 2019. Analysis was performed using Stata, release 15 (StataCorp, LLC) and SPSS Statistics for Windows, version 25.0 (IBM Corp). Basic descriptive statistics were used to describe distribution of characteristics in the total cohort and the groups of patients with complications. Stepwise multivariable logistic regression with clustering by institution was used to explore the association of variables with each of the study outcomes; α = .05 was used for selection of variables remaining in the models. To create a risk stratification system, conjunctive consolidation was performed by conjoining multiple independent variables.10 The strength of association between the risk stratification model and outcomes after controlling for other potential risk factors was assessed using odds ratios (ORs) and 95% CIs. We used the C statistic to assess the discrimination of the model including the new stratification system
Results
Seventeen academic centers contributed a total of 376 patients. Descriptive characteristics of the entire cohort are presented in Table 1. The range of patients from each institution was 3 to 39, with a median of 27. The median age of the patients was 83 years (range, 80-98 years); 253 patients (67.3%) were men and 123 (32.7%) were women. Before surgery, 306 patients (81.4%) lived independently, 52 (13.8%) were in an assisted living arrangement, and 18 (4.9%) were dependent. Body mass index (BMI; calculated as weight in kilograms divided by square of height in meters) ranged from 14.8 to 47.4 (median, 24.8). One hundred ten patients (29.3%) reported preoperative weight loss. Most procedures were performed for cancer.
Table 1. Distribution of Characteristics and Their Association With Study Outcomes.
| Factor | All Participants (N = 376) | Odds Ratio (95% CI) | ||
|---|---|---|---|---|
| 30-d Severe Complication (n = 193 [51%]) | 90-d Mortality (n = 30 [8%]) | 90-d Functional Decline (n = 36 [11%])a | ||
| Preoperative Factors | ||||
| Age, median (range), y | 83 (80-98) | 1.05 (0.99-1.11) | 1.20 (1.12-1.28) | 0.96 (0.84-1.10) |
| Sex, No. (%) | ||||
| Male | 253 (67.3) | 1.01 (0.76-1.33) | 1.67 (0.81-3.45) | 0.71 (0.31-1.63) |
| Female | 123 (32.7) | 1 [Reference] | 1 [Reference] | 1 [Reference] |
| BMI, median (range) | 24.9 (14.8-47.4) | 0.95 (0.92-0.99) | 0.96 (.92-1.00) | 0.91 (0.88-0.95) |
| Prior head and neck RT, No. (%) | ||||
| No | 232 (61.7) | 1 [Reference] | 1 [Reference] | 1 [Reference] |
| Yes | 144 (38.3) | 1.32 (0.82-2.12) | 2.26 (0.87-5.87) | 1.31 (0.54-3.17) |
| Smoking, No. (%) | ||||
| No | 169 (44.9) | 1 [Reference] | 1 [Reference] | 1 [Reference] |
| Yes | 207 (55.1) | 1.03 (0.78-1.36) | 0.80 (0.40-1.60) | 1.00 (0.52-1.93) |
| ACE-27 score, No. (%) | ||||
| 0-1 (none to mild) | 193 (51.3) | 1 [Reference] | 1 [Reference] | 1 [Reference] |
| 2-3 (moderate to severe) | 183 (48.7) | 1.83 (1.28-2.61) | 3.16 (1.15-8.68) | 0.84 (0.45-1.59) |
| Frailty, No. (%) | ||||
| <0.25 (low) | 243 (64.6) | 1 [Reference] | 1 [Reference] | 1 [Reference] |
| ≥0.25 (high) | 133 (35.4) | 1.66 (1.12-2.44) | 2.64 (0.85-8.20) | 1.55 (0.73-3.31) |
| ASA classification, No. (%)b | ||||
| 1-2 | 76 (20.2) | 1 [Reference] | 1 [Reference] | 1 [Reference] |
| 3-4 | 300 (79.8) | 1.22 (0.59-2.53) | 2.42 (0.88-6.71) | 1.85 (0.60-5.72) |
| Weight loss ≥4.5 kg, No. (%) | 110 (29.3) | 1.06 (0.75-1.51) | 1.55 (0.69-3.47) | 1.65 (0.69-3.96) |
| Functional status at baseline, No. (%) | ||||
| Independent | 306 (81.4) | 1 [Reference] | 1 [Reference] | 1 [Reference] |
| Dependent plus assisted living | 70 (18.6) | 1.54 (0.96-2.48) | 0.47 (0.14-1.53) | 1.70 (0.70-4.15) |
| Operative Factors | ||||
| Tumor site, No. (%) | ||||
| Scalp, facial skin, or parotid | 131 (34.8) | 1 [Reference] | 1 [Reference] | 1 [Reference] |
| Maxilla, oral cavity, or oropharynx | 207 (55.1) | 1.30 (0.79-2.14) | 0.81 (0.31-2.11) | 3.03 (1.59-5.78) |
| Larynx, hypopharynx, or esophagus | 38 (10.1) | 1.58 (0.93-2.69) | 2.71 (0.86-8.57) | 2.06 (0.52-8.16) |
| Flap type, No. (%) | ||||
| Free | 281 (74.7) | 1 [Reference] | 1 [Reference] | 1 [Reference] |
| Pedicle | 95 (25.3) | 0.64 (0.37-1.11) | 1.29 (0.60-2.78) | 0.71 (0.28-1.78) |
| Bony reconstruction, No. (%)c | ||||
| No | 214 (76.2) | 1 [Reference] | 1 [Reference] | 1 [Reference] |
| Yes | 67 (23.8) | 0.91 (0.49-1.67) | 0.99 (0.42-2.35) | 1.47 (0.63-3.43) |
| Duration of surgery, median (range), h | 8.7 (1.5-20.8) | 1.10 (1.2-1.19) | 1.03 (0.92-1.20) | 1.15 (1.00-1.33) |
| Additional operations, No. (%) | ||||
| No | 297 (79.0) | 1 [Reference] | 1 [Reference] | 1 [Reference] |
| Yes | 79 (21.0) | 5.1 (2.93-8.72) | 1.68 (0.78-3.60) | 2.80 (1.79-4.40) |
| Flap failure, No. (%) | ||||
| No | 350 (93.1) | 1 [Reference] | 1 [Reference] | 1 [Reference] |
| Partial plus total | 26 (6.9) | 2.75 (1.30-5.80) | 3.08 (1.21-7.79) | 1.47 (0.63-3.46) |
| LOS, median (range), d | 8 (2-101) | 1.15 (1.10-1.20) | 1.04 (1.01-1.08) | 1.08 (1.05-1.12) |
Abbreviations: ACE-27, Adult Comorbidity Evaluation–27; ASA, American Society of Anesthesiologists; BMI, body mass index (calculated as weight in kilograms divided by square of height in meters); LOS, length of stay; RT, radiotherapy.
Includes 328 patients who were not dependent at baseline and were alive at 90 days.
Higher scores indicate worsening overall health.
Includes 281 patients with free flaps.
In this cohort, 131 patients (34.8%) had surgery of the scalp, facial skin, or parotid; 207 (55.1%), the maxilla, oropharynx, or oral cavity; and 38 (10.1%), the larynx, hypopharynx, or esophagus. Two hundred eighty-one patients (74.7%) underwent free tissue transfer. One patient underwent pedicle and free flap reconstruction and was included in the free flap group. The most commonly performed free flap was anterolateral thigh (n = 114), and the most common pedicle flap was pectoralis (n = 65). Bony reconstruction was performed in 67 patients (23.8%) undergoing free flap reconstruction. The median length of stay was 8 days (range, 2-101 days). Overall 54 patients (14.4%) were readmitted within 30 days.
A comparison between the free flap and pedicle flap groups is presented in Table 2. There was a 24% higher prevalence of men (81 of 95 [85.3%] vs 172 of 281 [61.2%]; 95% CI, 14.8%-33.2%) and a 16% higher prevalence of prior head and neck radiotherapy (47 of 95 [49.5%] vs 96 of 281 [34.2%]; 95% CI, 4.5%-27.5%) in the pedicle flap group compared with the free flap group. Comorbidity and frailty metrics (ACE-27 score, American Society of Anesthesiologists classification, functional status, and MFI), BMI, and smoking history were similar between the groups. Median surgical time was greater in the free flap group (9.2 [range, 1.5-20.8] vs 5.9 [range, 2.2-13.8] hours; median difference, 3.2 hours; 95% CI, 2.5-3.8 hours). There were meaningful differences in the prevalence of ablation sites between the groups: a 10% higher prevalence of scalp, facial skin, or parotid sites in the pedicle flap group (95% CI, −21.3% to −1.3%), a 19% higher prevalence of maxilla, oral cavity, or oropharynx sites in the free flap group (95% CI, 7.6%-30.4%), and a 9% higher prevalence of larynx, hypopharynx, or esophagus sites in the pedicle flap group (95% CI, −17.2% to −0.8%).
Table 2. Distribution of Characteristics Between Groups Defined by Type of Flap.
| Factor | Flap Type | Difference (95% CI)a | |
|---|---|---|---|
| Free (n = 281) | Pedicle (n = 95) | ||
| Preoperative Factors | |||
| Age, median (range), y | 83 (80 to 95) | 84 (80 to 98) | −1 (−2 to 0) |
| Sex, No. (%) | |||
| Male | 172 (61.2) | 81 (85.3) | 24.0 (14.8 to 33.2) |
| Female | 109 (38.8) | 14 (14.7) | |
| BMI, median (range) | 24.7 (14.8 to 44.3) | 25.0 (16.5 to 47.4) | −0.5 (−1.5 to 0.6) |
| Prior head and neck RT, No. (%) | |||
| No | 185 (65.8) | 47 (49.5) | 16.0 (4.5 to 27.5) |
| Yes | 96 (34.2) | 48 (50.5) | |
| Smoking, No. (%) | |||
| No | 135 (48.0) | 34 (35.8) | 12.0 (0.7 to 23.3) |
| Current or ever | 146 (52.0) | 61 (64.2) | |
| ACE-27 score, No. (%) | |||
| 0-1 (none to mild) | 149 (53.0) | 44 (46.3) | 6.0 (−5.0 to 18.0) |
| 2-3 (moderate to severe) | 132 (47.0) | 51 (53.7) | |
| Frailty, No. (%) | |||
| <0.25 (low) | 183 (65.1) | 60 (63.2) | 2.0 (−9.2 to 13.2) |
| ≥0.25 (high) | 98 (34.9) | 35 (36.8) | |
| ASA classification, No. (%)b | |||
| 1-2 | 61 (21.7) | 15 (15.8) | 6.0 (−2.8 to 14.8) |
| 3-4 | 220 (78.3) | 80 (84.2) | |
| Weight loss ≥4.5 kg, No. (%) | 85 (30.21) | 25 (26.3) | 3.0 (−7.8 to 13.8) |
| Functional status, No. (%) | |||
| Independent | 227 (80.8) | 79 (83.2) | −2.0 (−10.8 to 6.8) |
| Dependent plus assisted living | 54 (19.2) | 16 (16.8) | |
| Operative Factors | |||
| Tumor site, No. (%) | |||
| Scalp, facial skin, or parotid | 91 (32.4) | 40 (42.1) | −10.0 (−21.3 to −1.3) |
| Maxilla, oral cavity, or oropharynx | 168 (59.8) | 39 (41.1) | 19.0 (7.6 to 30.4) |
| Larynx, hypopharynx, or esophagus | 22 (7.8) | 16 (16.8) | −9.0 (−17.2 to −0.8) |
| Duration of surgery, median (range), h | 9.2 (1.5 to 20.8) | 5.9 (2.2 to 13.8) | 3.2 (2.5 to 3.8) |
| Additional operations, No. (%) | |||
| No | 212 (75.4) | 85 (89.5) | 14.0 (5.9 to 22.1) |
| Yes | 69 (24.6) | 10 (10.5) | |
| Flap failure, No. (%) | |||
| No | 261 (92.9) | 89 (93.7) | 1.0 (−4.6 to 6.6) |
| Partial plus total | 20 (7.1) | 6 (6.3) | |
| Adjuvant therapy, No. (%) | |||
| No | 130 (50.0) | 48 (60.0) | 10.0 (−2.3 to 22.3) |
| Yes | 130 (50.0) | 32 (40.0) | |
| LOS, median (range), d | 9 (4 to 101) | 7 (2 to 45) | 2.3 (2.0 to 3.0) |
| 30-d readmission, No. (%) | 38 (13.5) | 16 (16.8) | −3.0 (−12 to 5) |
Abbreviations: ACE-27, Adult Comorbidity Evaluation–27; ASA, American Society of Anesthesiologists; BMI, body mass index (calculated as weight in kilograms divided by the square of height in meters); LOS, length of stay; OR, odds ratio; RT, radiotherapy.
For continuous level variables, difference is the median difference.
Higher scores indicate worsening overall health.
For the entire cohort, 13 patients had total flap loss and 13, partial flap loss (3.5% for both). Ten patients had total free flap loss (3.6%) and 3 had total pedicle flap loss (3.2%). Ten patients had partial free flap loss (3.6%) and 3 had partial pedicle flap loss (3.2%). There was no clinically meaningful difference between groups in overall flap failure rate (total and partial combined). Seventy-nine patients (21.0%) underwent at least 1 additional operation, of whom 19 had 2 or more additional operations. The number of patients undergoing additional operations was 14% higher in the free flap group compared with the pedicle flap group (69 of 281 [24.6%] vs 10 of 95 [10.5%]; 95% CI, 5.9%-22.1%). The median length of stay was 2.3 days longer in the free flap group (9 [range, 4-101] vs 7 [range, 2-45] days; 95% CI, 2.0-3.0 days). The 30-day readmission rate was similar between the groups (38 of 281 [13.5%] in the free flap group and 16 of 95 [16.8%] in the pedicle group).
30-Day Serious Complications
Thirty-day serious complications were reported in 193 patients (51.3%). The use of bony vs nonbony free flaps was not associated with this outcome in univariate analysis (OR, 0.91; 95% CI, 0.49-1.67) (Table 1). Rate of complications was 11% higher in the free flap group compared with the pedicle flap group (54% vs 43%; 95% CI, −0.5% to 22.5%). In multivariable analysis, moderate/severe comorbidity (ACE-27 score of 2-3 vs 0-1; OR, 1.80; 95% CI, 1.34-2.41), high frailty (MFI ≥0.25 vs <0.25; OR, 1.72; 95% CI, 1.10-2.67), and additional surgical procedures (OR, 5.40; 95% CI, 3.09-9.43) were associated with higher risk for complications. Higher BMI (≥25 vs <25) was protective (OR, 0.95; 95% CI, 0.91-0.99). After controlling for other variables in the model, pedicle vs free flap was not associated with 30-day complications (OR, 0.79; 95% CI, 0.40-1.58).
90-Day Mortality
Thirty patients died within 90 days of surgery (8.0%), including 21 (7.5%) in the free flap group and 9 (9.5%) in the pedicle flap group. Each additional year in age was associated with a 20% increased risk of mortality within 90 days (OR, 1.20; 95% CI, 1.12-1.28). The use of bony vs nonbony free flap was not associated with this outcome in univariate analysis (OR, 0.99; 95% CI, 0.42-2.35) (Table 1). In multivariable analysis, age (≥85 vs <85 years; OR, 1.19; 95% CI, 1.14-1.26), moderate/severe comorbidity (ACE-27 score 2-3 vs 0-1; OR, 3.33; 95% CI, 1.29-8.60), and flap failure (total or partial vs none; OR, 3.56; 95% CI, 1.47-8.62) were associated with higher risk for 90-day mortality. After controlling for other variables in the model, pedicle vs free flap was not associated with 90-day mortality (OR, 0.98; 95% CI, 0.41-2.35).
90-Day Functional Decline
From the cohort, 328 patients who were not dependent at baseline were alive at 90 days and included in this analysis. Of these patients, 36 (11.0%) experienced a decline to dependent status at 90 days, including 29 of 247 patients (11.7%) in the free flap group and 7 of 81 (8.6%) in the pedicle group (difference, 3.1%; 95% CI, −4.2% to 10.5%). The use of bony vs nonbony free flap was not associated with this outcome in univariate analysis (OR, 1.47; 95% CI, 0.63-3.43) (Table 1). In multivariable analysis, maxillary, oral cavity, or oropharynx sites (OR, 2.51; 95% CI, 1.30-4.85) and any additional surgical procedures (OR, 2.94; 95% CI, 1.81-4.79) were associated with higher risk for 90-day functional decline. After controlling for other variables in the model, pedicle vs free flap was not associated with 90-day functional decline (OR, 1.66; 95% CI, 0.57-4.83).
Risk Stratification System
The following clinically important preoperative patient variables were conjoined using conjunctive consolidation to create a risk stratification system: age (<85 vs ≥85 years), comorbidity (ACE-27 scores 0-1 vs 2-3), BMI (<25 vs ≥25), and frailty (MFI<0.25 vs ≥0.25). Conjunctive consolidation was performed first for 30-day serious complications. Age was conjoined with comorbidity to create a 3-category staging system. This staging system was then conjoined with BMI to create a second 3-category staging system. The second system was then conjoined with frailty, resulting in a third and final 3-tier risk stratification system, which demonstrated excellent gradients for 30-day complications (Table 3 and Table 4). The conjunctive consolidation process resulted in the same classification system for 90-day mortality. Thirty-day serious complications occurred in 85 of 198 patients (42.9%) for class I, 69 of 123 (56.1%) for class II, and 39 of 55 (70.9%) for class III. Ninety-day mortality occurred in 7 of 198 (3.5%) for class I, 13 of 122 (10.7%) for class II, and 10 of 54 (18.5%) for class III. Ninety-day functional decline with the new classification system occurred in 19 of 186 (10.2%) for class I, 10 of 105 (9.5%) for class II, and 7 of 37 (18.9%) for class III.
Table 3. Risk Classification System Based on Preoperative Patient Factors.
| Age, y | Comorbiditya | BMI | Frailtyb | Classification |
|---|---|---|---|---|
| <85 | 0-1 | ≥25 | Low | I |
| High | I | |||
| <25 | Low | I | ||
| High | II | |||
| 2-3 | ≥25 | Low | I | |
| High | II | |||
| <25 | Low | II | ||
| High | III | |||
| ≥85 | 0-1 | ≥25 | Low | I |
| High | I | |||
| <25 | Low | I | ||
| High | II | |||
| 2-3 | ≥25 | Low | II | |
| High | III | |||
| <25 | Low | II | ||
| High | III |
Abbreviation: BMI, body mass index (calculated as weight in kilograms divided by height in meters squared).
Calculated using the Adult Comorbidity Evaluation–27 score, with 0 to 1 indicating none to mild comorbidities and 2 to 3 indicating moderate to severe comorbidities
Calculated using the Modified Frailty Index, with a maximum of 11 points, and 0 to 2 points indicating low frailty and 3 or more points indicating high frailty (or ≥0.25).
Table 4. Outcomes of Risk Classification System.
| Risk Classification System | Patients, No./Total No. (%) | ||
|---|---|---|---|
| 30-d Serious Complications | 90-d Mortality | 90-d Functional Decline | |
| Class I | 85/198 (42.9) | 7/198 (3.5) | 19/186 (10.2) |
| Class II | 69/123 (56.1) | 13/122 (10.7) | 10/105 (9.5) |
| Class III | 39/55 (70.9) | 10/54 (18.5) | 7/37 (18.9) |
| C statistic | 0.72 | 0.72 | 0.70 |
After controlling for other risk factors, patients in classes II and III had higher risk for 30-day complications and 90-day mortality as compared with patients in class I (Table 5). Only class III patients were at higher risk for 90-day functional decline after controlling for other risk factors. Multivariable models containing the new risk classification system showed good discriminative power for each of the outcomes, with C statistics of 0.72 for 30-day complication, 0.72 for 90-day mortality and 0.70 for functional decline. When using the new risk classification in multivariable models, duration of surgery remained an independent risk factor in this analysis (OR, 1.08; 95% CI, 1.00-1.16) (Table 5).
Table 5. Multivariable Model With Risk Classification System.
| Variable | Odds Ratio (95% CI) | ||
|---|---|---|---|
| 30-d Severe Complicationa | 90-d Mortalityb | 90-d Functional Declinec,d | |
| Risk classification system | |||
| Class I | 1 [Reference] | 1 [Reference] | 1 [Reference] |
| Class II | 1.87 (1.28-2.72) | 3.39 (1.22-9.42) | NA |
| Class III | 4.50 (2.44-8.28) | 6.67 (1.40-31.85) | 2.49 (1.11-5.57) |
| Additional operations | |||
| No | 1 [Reference] | 1 [Reference] | 1 [Reference] |
| Yes | 5.04 (2.59-9.84) | NA | 3.24 (1.79-5.86) |
| Duration of surgery | 1.08 (1.00-1.16) | NA | NA |
| Flap failure | NA | 3.63 (1.48-8.89) | NA |
| Tumor site | |||
| Scalp, facial skin, or parotid | NA | NA | 1 [Reference] |
| Maxilla, oral cavity, or oropharynx | NA | NA | 3.07 (1.51-6.25) |
| Larynx, hypopharynx, or esophagus | NA | NA | NA |
Abbreviation: NA, not applicable.
Additional variables tested in multivariable model include ablation site, American Society of Anesthesiologists (ASA) classification, flap failure, and flap type.
Additional variables tested in multivariable model include additional operations, ASA classification, and flap type.
Additional variables tested in multivariable model include additional operations, ASA classification, flap failure, flap type, prior head and neck radiotherapy, and duration of surgery.
Includes 328 patients who were not dependent at baseline and were alive at 90 days.
Discussion
This large, retrospective cohort study represents a major contribution to the literature by combining the multi-institutional experience of patients 80 years or older undergoing major head and neck surgery, as defined by reconstruction in the form of a free flap or a pedicle flap. Within this cohort we found that more than half of these elderly patients had a serious postoperative complication (193 [51.3%]), 30 (8.0%) died within 90 days of surgery, and 36 of those who were not dependent at baseline (11.0%) declined to dependent status at 90 days. These findings are similar to those of other studies that have reported medical complication rates in octogenarians undergoing microvascular reconstruction of the head and neck at 40% to 62%.14,15
A major objective of this study was to identify patient factors associated with postoperative outcomes. Most researchers and physicians agree that age alone is not an adequate risk factor for outcomes. We propose that frailty and comorbidity provide meaningful fine-tuning of the evaluation of the elderly patient. Frailty is defined as a “state of increased vulnerability, resulting from aging-associated decline in reserve and function across multiple physiologic systems such that the ability to cope with every day or acute stressors is compromised.”16(p1) We found that MFI of at least 0.25 was associated with higher risk of complications. Using the ACE-27 instrument, severe comorbidity has been shown to be associated with mortality in patients 80 years or older undergoing any type head and neck surgery.17 Similarly, we found ACE-27 scores of 2 or 3 (moderate to severe comorbidity) to be associated with 30-day serious complications and 90-day mortality in this cohort. Of note, American Society of Anesthesiologists classification was not associated with outcomes in this study.
Interestingly, high BMI (≥25) was found to be independently protective for 30-day serious complications. This finding is in line with those of other studies showing that higher BMI was associated with lower mortality in the elderly.18,19 We hypothesize that patients with a higher BMI have more reserve to endure the extensive surgery and recuperative period.
Based on these data, we designed a novel risk stratification system using preoperative patient factors. During preoperative counseling, a patient’s age, comorbidity, BMI, and frailty can be used to assign them into 1 of 3 classes. The associated risks of 30-day serious complications, 90-day mortality, and 90-day functional decline based on the risk class can be provided to the patient and family. Although this approach requires the calculation of ACE-27 scores and MFI during the clinic visit, such information can provide important risk assessment information for this select patient population. Once these instruments are familiar, the ACE-27 score and MFI can be quickly and easily calculated by physicians, fellows, residents, nurses, or even staff personnel.
Another objective of this study was to explore the association between the type of reconstructive surgery and outcomes. Within this study population of patients 80 years or older, we demonstrated that the type of flap (free vs pedicle) was not independently associated with higher risk of 30-day serious complications, 90-day mortality, or 90-day functional decline. The free flap failure rate among elderly patients in this study was 3.6%, which is similar to published rates for younger patients.20 Bony free flaps compared with soft tissue free flaps were also not associated with worse outcomes. These data support that free tissue reconstruction (with or without bone) need not be avoided in elderly patients if it is believed to provide superior reconstruction compared with a pedicle flap. These data also caution that additional operations, longer surgical times, or flap failure are associated with worse outcomes in this elderly population. Thus, the anticipated risks of additional operations, longer surgical times, or flap failure specific to a particular case should be the main considerations during counseling and surgical planning.
Patients in whom the ablative site was the maxilla, oral cavity, or oropharynx experienced a greater 90-day functional decline. We hypothesize that this may be owing to higher rates of postoperative dysphagia and aspiration compared with other surgical sites, resulting in greater debilitation and functional decline. This group may also have a higher chance of prolonged tracheostomy and/or tube feedings, both of which could factor into worse functional status.
Clinical experience suggests that elderly patients are more likely to decline curative surgical treatment if there is a high risk of postoperative morbidity. This study brings awareness to the importance of discussing functional decline in the preoperative setting by demonstrating a 10% to 20% rate of patients declining to a dependent status at 90 days.
Limitations
Certain limitations to this study must be acknowledged. First is the retrospective nature of the study and data collection. Another limitation is the inherent bias toward healthier, elderly patients who were deemed to be surgical candidates in the first place. We obviously cannot comment on patients who were believed to be too sickly and were not given the option for surgery or who did not elect to have surgery. This is evidenced by the fact that only 4.9% of the cohort was dependent at baseline. Another bias may be institutional preference toward pedicle or free flaps. The higher prevalence of prior head and neck radiotherapy in the pedicle group may suggest a tendency toward pedicle flap reconstruction in elderly patients who have had prior radiation. We attempted to minimize bias by including pedicle and free flap reconstruction and by providing comprehensive comorbidity and frailty data. In addition, all centers certified inclusion of all eligible patients within the study period, which minimized omissions that could introduce additional bias. Last, this study could not account for oncologic prognosis, which should be factored into the ideal risk prediction model.
Conclusions
In this multi-institutional, retrospective cohort study of patients 80 years and older undergoing major head and neck surgery who received free tissue transfer or pedicle flap reconstruction, we describe rates of 30-day complications, 90-day mortality, and 90-day functional decline. Being 85 years or older, moderate or severe comorbidities, BMI of less than 25, high frailty, surgical hours, flap failure, additional operations, and surgery of the maxilla, oral cavity or oropharynx were associated with worse outcomes. Neither free tissue transfer nor bony reconstruction were associated with worse outcomes. A novel risk classification system based on age, BMI, comorbidity, and frailty is presented to assist in preoperative patient counseling and surgical planning.
References
- 1.Population Reference Bureau Fact sheet: aging in the United States. https://www.prb.org/aging-unitedstates-fact-sheet. Accessed March 1, 2019.
- 2.American Cancer Society Cancer in the oldest old. Cancer Facts and Figures 2019. https://www.cancer.org/content/dam/cancer-org/research/cancer-facts-and-statistics/annual-cancer-facts-and-figures/2019/cancer-facts-and-figures-special-section-cancer-in-the-oldest-old-2019.pdf. Accessed March 1, 2019.
- 3.Smith BD, Smith GL, Hurria A, Hortobagyi GN, Buchholz TA. Future of cancer incidence in the United States: burdens upon an aging, changing nation. J Clin Oncol. 2009;27(17):2758-2765. doi: 10.1200/JCO.2008.20.8983 [DOI] [PubMed] [Google Scholar]
- 4.Massa ST, Cass L, Challapalli S, et al. Age differences in demographic predictors of head and neck cancer survival. Cancer Epidemiol Biomarkers Prev. 2018;27(7):A88. doi: 10.1158/1538-7755.DISP17-A88 [DOI] [Google Scholar]
- 5.Mitchell CA, Goldman RA, Curry JM, et al. Morbidity and survival in elderly patients undergoing free flap reconstruction: a retrospective cohort study. Otolaryngol Head Neck Surg. 2017;157(1):42-47. doi: 10.1177/0194599817696301 [DOI] [PubMed] [Google Scholar]
- 6.Beausang ES, Ang EE, Lipa JE, et al. Microvascular free tissue transfer in elderly patients: the Toronto experience. Head Neck. 2003;25(7):549-553. doi: 10.1002/hed.10240 [DOI] [PubMed] [Google Scholar]
- 7.Tarsitano A, Pizzigallo A, Sgarzani R, Oranges CM, Cipriani R, Marchetti C. Head and neck cancer in elderly patients: is microsurgical free-tissue transfer a safe procedure? Acta Otorhinolaryngol Ital. 2012;32(6):371-375. [PMC free article] [PubMed] [Google Scholar]
- 8.Sierakowski A, Nawar A, Parker M, Mathur B. Free flap surgery in the elderly: experience with 110 cases aged ≥70 years. J Plast Reconstr Aesthet Surg. 2017;70(2):189-195. doi: 10.1016/j.bjps.2016.11.008 [DOI] [PubMed] [Google Scholar]
- 9.Velanovich V, Antoine H, Swartz A, Peters D, Rubinfeld I. Accumulating deficits model of frailty and postoperative mortality and morbidity: its application to a national database. J Surg Res. 2013;183(1):104-110. doi: 10.1016/j.jss.2013.01.021 [DOI] [PubMed] [Google Scholar]
- 10.Piccirillo JF, Tierney RM, Costas I, Grove L, Spitznagel EL Jr. Prognostic importance of comorbidity in a hospital-based cancer registry. JAMA. 2004;291(20):2441-2447. doi: 10.1001/jama.291.20.2441 [DOI] [PubMed] [Google Scholar]
- 11.Mogal H, Vermilion SA, Dodson R, et al. Modified Frailty Index predicts morbidity and mortality after pancreaticoduodenectomy. Ann Surg Oncol. 2017;24(6):1714-1721. doi: 10.1245/s10434-016-5715-0 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Tsiouris A, Hammoud ZT, Velanovich V, Hodari A, Borgi J, Rubinfeld I. A modified frailty index to assess morbidity and mortality after lobectomy. J Surg Res. 2013;183(1):40-46. doi: 10.1016/j.jss.2012.11.059 [DOI] [PubMed] [Google Scholar]
- 13.Ali R, Schwalb JM, Nerenz DR, Antoine HJ, Rubinfeld I. Use of the modified frailty index to predict 30-day morbidity and mortality from spine surgery. J Neurosurg Spine. 2016;25(4):537-541. doi: 10.3171/2015.10.SPINE14582 [DOI] [PubMed] [Google Scholar]
- 14.Howard MA, Cordeiro PG, Disa J, et al. Free tissue transfer in the elderly: incidence of perioperative complications following microsurgical reconstruction of 197 septuagenarians and octogenarians. Plast Reconstr Surg. 2005;116(6):1659-1668. doi: 10.1097/01.prs.0000187135.49423.9f [DOI] [PubMed] [Google Scholar]
- 15.Blackwell KE, Azizzadeh B, Ayala C, Rawnsley JD. Octogenarian free flap reconstruction: complications and cost of therapy. Otolaryngol Head Neck Surg. 2002;126(3):301-306. doi: 10.1067/mhn.2002.122704 [DOI] [PubMed] [Google Scholar]
- 16.Xue QL. The frailty syndrome: definition and natural history. Clin Geriatr Med. 2011;27(1):1-15. doi: 10.1016/j.cger.2010.08.009 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.L’Esperance HE, Kallogjeri D, Yousaf S, Piccirillo JF, Rich JT. Prediction of mortality and morbidity in head and neck cancer patients 80 years of age and older undergoing surgery. Laryngoscope. 2018;128(4):871-877. doi: 10.1002/lary.26858 [DOI] [PubMed] [Google Scholar]
- 18.Kvamme JM, Holmen J, Wilsgaard T, Florholmen J, Midthjell K, Jacobsen BK. Body mass index and mortality in elderly men and women: the Tromso and HUNT studies. J Epidemiol Community Health. 2012;66(7):611-617. doi: 10.1136/jech.2010.123232 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Janssen I, Katzmarzyk PT, Ross R. Body mass index is inversely related to mortality in older people after adjustment for waist circumference. J Am Geriatr Soc. 2005;53(12):2112-2118. doi: 10.1111/j.1532-5415.2005.00505.x [DOI] [PubMed] [Google Scholar]
- 20.Wu CC, Lin PY, Chew KY, Kuo YR. Free tissue transfers in head and neck reconstruction: complications, outcomes and strategies for management of flap failure: analysis of 2019 flaps in single institute. Microsurgery. 2014;34(5):339-344. doi: 10.1002/micr.22212 [DOI] [PubMed] [Google Scholar]