Aspiration pneumonia after concurrent chemoradiotherapy for head-and-neck cancer

Abstract

Background

Aspiration pneumonia represents an under-reported complication of chemoradiotherapy in head-and-neck cancer. This study evaluated the incidence, risk factors, and mortality of aspiration pneumonia in a large cohort of head-and-neck cancer patients treated with concurrent chemoradiotherapy.

Methods

Patients with head-and-neck cancer diagnosed between 2000 and 2009 were identified from the SEER-Medicare database. Aspiration pneumonia was identified from Medicare billing claims. The cumulative incidence, risk factors, and survival after aspiration pneumonia were estimated and compared to a non-cancer population.

Results

Of 3,513 head-and-neck cancer patients, 801 patients developed aspiration pneumonia at a median time of 5 months after initiating treatment. The 1- and 5-year cumulative incidence of aspiration pneumonia was 15.8% and 23.8% for head-and-neck cancer patients and 3.6% and 8.7% for non-cancer controls, respectively. Among cancer patients multivariate analysis identified independent risk factors (p<0.05) for aspiration pneumonia including hypopharyngeal and nasopharyngeal tumors, male gender, older age, increased comorbidity, no surgery prior to radiation, and care received at a teaching hospital. Among cancer patients who experienced aspiration pneumonia, 674 (84%) were hospitalized of which 301 (45%) were admitted to an intensive care unit. Thirty-day mortality after hospitalization for aspiration pneumonia was 32.5%. Aspiration pneumonia was associated with a 42% increased risk of death (HR=1.42, p<0.001) after controlling for confounders.

Conclusions

This study found that nearly one-quarter of elderly patients will develop aspiration pneumonia within 5 years of chemoradiotherapy for head-and-neck cancer. A better understanding of mitigating factors will help identify patients at risk for this potentially lethal complication.

Introduction

Chemotherapy combined with radiation represents a standard treatment approach for locally advanced head-and-neck cancer. While this joint treatment modality has a proven survival benefit, it also poses the risk of significant acute and late toxicities that can have a profound impact on survivorship and quality of life of among head-and-neck cancer survivors.

Aspiration pneumonia―defined as pneumonia secondary to inhalation of food particles, saliva, or other foreign substances―represents an under-reported side effect of head-and-neck cancer after treatment with chemoradiotherapy^{1, 2}. Aspiration pneumonia after radiation is likely due to a combination of contributing factors including acute and chronic radiation induced mucosal changes, muscle fibrosis, and xerostomia ^{1, 3–6}. These factors lead to swallowing dysfunction which in turn increases the risk of both aspiration, and aspiration pneumonia.

Previous studies characterizing the risk of swallowing dysfunction after radiation have identified dysphagia or aspiration rates ranging from 33 to 81% ^{4, 6–8}. Additional research has shown that aspiration pneumonia is a major source of post-treatment morbidity as well as a potential cause of death among head-and-neck cancer patients ^{1, 9}. The current body of literature on aspiration pneumonia includes single institution analyses ^7–12 often with small sample sizes ^7–11 or limited follow-up ^{13, 14}. To date, population-based studies that characterize the risk of aspiration pneumonia after chemoradiation do not exist. The purpose of this study was to use SEER-Medicare linked data to evaluate the incidence, risk factors, morbidity, and mortality of aspiration pneumonia in a large cohort of head-and-neck cancer patients treated with concurrent chemoradiotherapy.

Methods

Data source

This study evaluated head-and-neck cancer patients within the Surveillance, Epidemiology, and End Results (SEER)-Medicare linked database. The SEER program consists of a collection of cancer registries geographically spread across the US which collect demographic, clinical, treatment, and survival information for individuals with cancer ¹⁵. The Medicare program provides federally funded health insurance for people over the age of 65. The SEER-Medicare linkage combines longitudinal Medicare claims data for patients within the SEER database providing a valuable resource to understand patterns of care and health outcomes for cancer patients from before diagnosis to throughout treatment with complete follow-up extending through death.

Aspiration pneumonia represents a known complication among older adults without cancer; therefore this study compared aspiration pneumonia rates among head-and-neck cancer patients to that of non-cancer controls. The non-cancer patients were drawn from a 5% random sample of Medicare beneficiaries residing in the same geographic regions covered by SEER. The contents of Medicare files for the non-cancer control group were identical to those available for patients with cancer ¹⁵. This study was found to be exempt from approval by the Institutional Review Board at the University of California San Diego.

Study population

The initial query of the SEER-Medicare database identified 42,250 patients with head-and-neck cancer over the age of 66 diagnosed between 2000 through 2009. Even though Medicare coverage starts at 65, only patients over 66 were included to ensure a complete year of Medicare claims data to enable calculation of Charlson comorbidity scores (described below). Patients with invalid dates of diagnosis (n=297) and those with non-squamous cell carcinoma histology (n=9,326) were excluded. The study cohort was further refined by removing patients diagnosed with multiple primary tumors (n=8,529) and those diagnosed at autopsy or death (n = 202). Patients with missing data were excluded (n=8,642) including those with non-continuous part A and part B coverage or any part C enrollment (enrollment in an HMO) from 12 months prior to diagnosis to death or last follow-up, leaving 15,254 remaining patients. Finally, only patients who received radiation with concurrent chemotherapy within 6 months of diagnosis were included leaving 3,513 cancer patients in the final study cohort.

Study covariates

Demographic covariates obtained from SEER included patient age at diagnosis, gender, race, and marital status. Geographic location was based on the location of SEER cancer registries, which were categorized into East (Connecticut, New Jersey), Midwest (Detroit, Iowa), South (Atlanta, Rural Georgia, Kentucky, Louisiana), and West (San Francisco, Hawaii, New Mexico, Seattle, Utah, San Jose, Los Angeles, Greater California). Median regional household income was determined from year 2000 Census tract data and divided into quartiles. Patients without household income data (1%) were grouped into the bottom quartile. Tumor site was categorized into six groups: oral cavity, nasopharynx, oropharynx and tonsil, larynx, hypopharynx, and other (salivary gland, other oral cavity and pharynx, and nose, nasal cavity, and middle ear). The SEER Historic Stage A staging system¹⁶ was used to classify cancer stage as TNM staging data was limited for the years studied and defined differently across head-and-neck cancer sites. Patients staged as “distant” in the historic SEER system for head-and-neck cancer have locally advanced disease equivalent to T4a and T4b in the American Joint Committee on Cancer (AJCC) staging system¹⁷.

Comorbidity score was calculated during the year prior to diagnosis using inpatient and outpatient Medicare claims with the Deyo modification of the Charlson comorbidity index ¹⁸. Patient care within a teaching hospital was defined as any indirect medical education charge on a Medicare claim during a hospitalization after the patient’s diagnosis of cancer.

Treatment covariates included surgery, radiation, and chemotherapy. Information on whether the patient had surgery directed at the primary tumor was obtained from SEER and Medicare billing claims (Table 1). The delivery of radiation and chemotherapy were determined from Medicare billing claims (Table 1) using methods as described previously ^{19, 20}. Concurrent chemotherapy was defined as a chemotherapy claim anytime from 2 weeks before the start of radiation through 2 weeks after completing radiation. Additionally, we assessed for the use of additional chemotherapy before (neoadjuvant), or after (adjuvant) the patient received radiotherapy. Use of positron emission tomography (PET) was defined from Medicare claims from 3 months prior to diagnosis extending through the start of radiation. Intensity-modulated radiation therapy (IMRT) was defined as the existence of any IMRT billing claim during the course of radiation. Table 2 summarizes the baseline characteristics of the 3,513 head-and-neck cancer patients included in this study.

Table 1.

Medicare and SEER codes used in study

Variable	Codes
Radiotherapy	HCPCS codes: 61796–61800, 63620–63621, 77371–77373, 77401–77416, 77418, 77421–77423, 77470, 77520, 77522, 77523, 77525, 0197T, G0173-G0174, G0243, G0251, and G0339-G0340.
Surgery	ICD-9 procedure codes: 21.5–21.6, 22.31, 22.42, 22.60–22.66, 24.31, 25.1–25.4, 26.2, 26.29–26.32, 27.3, 27.32, 27.4, 27.42–27.43, 27.49, 27.72, 28.92, 29.33, 29.39, 30.0, 30.09, 30.1, 30.21–30.22, 30.29, 30.3–30.5, 31.5, 40.40–40.42, 76.2, 76.31, 76.39–76.42, or 76.44–76.45. HCPCS codes: 21044–21045, 21555–21557, 30117–30118, 30130, 30140, 30150, 31200–31201, 31205, 31225, 31230, 31299, 31365, 31367–31368, 31370, 31375, 31380, 31382, 31390, 31395, 31420, 38700, 38720, 38724, 40810, 40812, 40814, 40816, 40819, 41110, 41112–41114, 41116, 41120, 41130, 41135, 41140, 41145, 41150, 41153, 41155, 41825–41827, 42104, 42106–42107, 42120, 42140, 42410, 42415, 42420, 42425–42456, 42440, 42450, 42842, 42844–42845, or 42890.
Chemotherapy	ICD-9 procedure code: 99.25. ICD-9 diagnosis codes: V58.1, V58.11, V58.12, V66.2, or V67.2. HCPCS codes: 96400 to 96599, J8999 to J9999, J8520, J8521, or Q0083 to Q0085. Revenue Center codes: 0331, 0332, or 0335. NDC: 00004110013, 00004110020, 00004110051, 00004110113, 00004110116, 00004110150, 00004110151. BETOS codes: 01D.
Aspiration pneumonia	ICD-9 diagnosis code: 507.0.
IMRT	HCPCS codes: 77301, 77338, 77418, G0174, G0178.
PET	HCPCS codes: 78810–78816, G0223-G0225.

Abbreviations: HCPCS, Healthcare Common Procedure Coding System codes; CPT, Current Procedural Terminology; NDC, National Drugs Codes; BETOS, Berenson-Eggers Type of Service; IMRT, Intensity-Modulated Radiation Therapy; PET, Positron Emission Tomography.

Table 2.

Descriptive characteristics of the study patients

Variables	N (%)
Total	3,513 (100)
Age at diagnosis, y
66–74	2,255 (64)
75–79	691 (20)
≥80	567 (16)
Metropolitan area
Yes	2,920 (83)
No	593 (17)
Teaching hospital
Yes	2,013 (57)
No	1,500 (43)
Region
East	776 (22)
Midwest	406 (12)
South	959 (27)
West	1372 (39)
Sex
Male	2,471 (70)
Female	1,042 (30)
Race
White	2,961 (84)
Black	329 (9)
Other	223 (6)
Marital status
Married	1,894 (54)
Other	1,619 (46)
Charlson comorbidity score
0	1,861 (53)
1	947(27)
2	395 (11)
≥3	310 (9)
Year of diagnosis
2000–2003	988 (28)
2004–2006	1,109 (32)
2007–2009	1,416 (43)
Tumor site
Hypopharynx	391 (11)
Larynx	834 (24)
Nasopharynx	148 (4)
Oral cavity	583 (17)
Oropharynx	1,298 (37)
Other	259 (7)
Historic stage
Localized	478 (14)
Regional	2,219 (63)
Distant	695 (20)
Unknown	121 (3)
PET
Yes	1,800 (51)
No	1,713 (49)
IMRT
Yes	1,821 (52)
No	1,692 (48)
Surgery prior to radiation
Yes	753 (21)
No	2,760 (79)
Sequence of chemotherapy and radiotherapy
During only	1,982 (56)
During, and before	458 (13)
During, and after	894 (25)
During, before, and after	179 (5)
Types of chemotherapy
Cisplatin	768 (22)
Cetuximab	588 (17)
Other	2,157 (61)

Abbreviations: PET, Positron Emission Tomography; IMRT, Intensity-Modulated Radiation Therapy.

Aspiration pneumonia

Aspiration pneumonia, the primary endpoint of this study, was identified from ICD-9 diagnosis code of 507.0 (Table 1) which represents pneumonitis or pneumonia due to inhalation of food or vomitus ²¹. The ICD-9 code for aspiration pneumonia is distinct from codes that describe bacterial pneumonia. A study attempting to determine the validity of this ICD-9 code found that 91% of patients with the ICD-9 code for aspiration pneumonia had objective clinical evidence or a physician note documenting aspiration pneumonia ²². In this study, aspiration pneumonia was defined as the presence of this ICD-9 diagnosis code within inpatient or outpatient Medicare billing claims any time after the start of radiation until the end of the study period, December 31, 2009, or the patient’s date of death. Aspiration pneumonia associated with hospitalization was defined as any aspiration pneumonia diagnosis code during a hospitalization. Similarly, admission to an intensive care unit (ICU) was defined as any aspiration pneumonia diagnosis code during admission to an ICU.

Statistical analysis

The rate of aspiration pneumonia was estimated with a competing risk analysis to account for the competing event of death. Time to an aspiration pneumonia event was defined from the start of radiation through the first aspiration pneumonia claim, censoring at last follow-up and treating death of any cause as a competing event. Unadjusted estimates of the rate of aspiration pneumonia were determined with nonparametric cumulative incidence functions using Gray’s test to compare differences between head-and-neck cancer patients and non-cancer controls as well as between subgroups among head-and-neck cancer patients ²³. Multivariate predictors of aspiration pneumonia were assessed with Fine-Gray regression models to estimate the sub-distribution hazards ^{24, 25}. All covariates were defined a priori based on factors that could impact survival or radiation therapy treatment. All variables included in the multivariate model are noted in Table 3.

Table 3.

Risk factors for aspiration pneumonia among patients with head-and-neck cancer treated with chemoradiotherapy.

	Univariate analysis		Multivariate analysis
Variables	5-yr rate (%)	SDHR (95% CI)	SDHR (95% CI)
Age at diagnosis, y
66–74	21.4	Ref	Ref
75–79	27.9	1.40 (1.17–1.67)	1.37 (1.14–1.64)
≥80	28.9	1.45 (1.20–1.75)	1.49 (1.22–1.82)
Metropolitan area
Yes	25.4	1.64 (1.31–2.06)	1.22 (0.95–1.57)
No	16.7	Ref	Ref
Teaching hospital
Yes	28.3	1.79 (1.53–2.09)	1.87 (1.58–2.20)
No	17.5	Ref	Ref
Region
West	25.3	Ref	Ref
East	26.6	1.06 (0.89–1.27)	0.89 (0.73–1.08)
Midwest	26.6	1.04 (0.82–1.30)	0.86 (0.67–1.09)
South	18.7	0.67 (0.55–0.81)	0.75 (0.60–0.92)
Median Income
Bottom quartile	19.9	Ref	Ref
2nd quartile	23	1.17 (0.94–1.46)	1.24 (0.98–1.55)
3rd quartile	25.2	1.31 (1.06–1.62)	1.25 (1.00–1.57)
Top quartile	27.9	1.54 (1.25–1.89)	1.37 (1.09–1.73)
Sex
Male	25.4	1.27 (1.08–1.50)	1.31 (1.09–1.56)
Female	20.2	Ref	Ref
Race
White	23.5	Ref	Ref
Black	24.8	1.01 (0.79–1.29)	1.07 (0.82–1.41)
Other	28.8	1.31 (1.01–1.71)	1.06 (0.80–1.41)
Marital status
Married	24	1.01 (0.88–1.17)	0.94 (0.81–1.10)
Other	23.8	Ref	Ref
Charlson comorbidity score
0	22.1	Ref	Ref
1	23.9	1.09 (0.92–1.29)	1.07 (0.90–1.27)
2	29.1	1.27 (1.02–1.59)	1.26 (1.01–1.58)
≥3	NA	1.15 (0.89–1.49)	1.13 (0.87–1.48)
Year of diagnosis
2000–2003	22.9	Ref	Ref ()
2004–2006	25	1.10 (0.92–1.31)	1.06 (0.87–1.30)
2007–2009	NA	0.89 (0.74–1.06)	0.86 (0.67–1.11)
Tumor site
Oral cavity	19.9	Ref	Ref
Hypopharynx	30.5	1.68 (1.28–2.20)	1.58 (1.20–2.10)
Larynx	24.4	1.20 (0.94–1.54)	1.16 (0.90–1.49)
Nasopharynx	34.1	1.76 (1.23–2.50)	1.77 (1.23–2.55)
Oropharynx	23.4	1.16 (0.92–1.46)	1.16 (0.91–1.47)
Other	19.9	0.96 (0.68–1.36)	0.94 (0.67–1.34)
Historic stage
Localized	23.8	Ref	Ref
Regional	23.5	1.03 (0.83–1.28)	0.99 (0.79–1.25)
Distant	25.2	1.07 (0.83–1.38)	1.01 (0.77–1.31)
Unknown	29.4	1.29 (0.86–1.94)	1.27 (0.84–1.92)
PET
Yes	24.5	1.00 (0.86–1.15)	0.99 (0.84–1.17)
No	23.6	Ref	Ref
IMRT
Yes	24.1	1.00 (0.87–1.16)	1.06 (0.88–1.26)
No	23.3	Ref	Ref
Surgery prior to raditaion
Yes	20.7	0.83 (0.69–1.00)	0.81 (0.67–0.99)
No	24.6	Ref	Ref
Sequence of chemotherapy and radiotherapy
During	24.8	Ref	Ref
During and before	21.6	0.85 (0.67–1.08)	0.86 (0.67–1.11)
During and after	23.9	0.96 (0.81–1.14)	0.94 (0.79–1.11)
During, before and after	21.4	0.89 (0.63–1.26)	0.86 (0.61–1.22)
Types of chemotherapy
Cetuximab	27.7	Ref	Ref
Cisplatin	23.9	0.86 (0.69–1.09)	0.89 (0.69–1.14)
Other	23.3	0.91 (0.75–1.11)	0.92 (0.73–1.15)

Abbreviations: SDHR, Subdistribution Hazard Ratio; PET, Positron Emission Tomography; IMRT, Intensity-Modulated Radiation Therapy.

We identified non-cancer control patients from the Medicare 5% sample using a 1:1 match for non-cancer patients with continuous Medicare parts A and B without part C for the entire observation period of 2000–2009 ²⁶. The match was based on patient characteristics (gender, income, and registry location), as well as age, survival status, and date of death. The time at risk for aspiration pneumonia in the head-and-neck cancer patients extended from the start of radiation through death or last follow-up. Non-cancer controls did not receive radiation therefore a pseudo-radiation start date was assigned for controls to ensure an identical at-risk time period to evaluate for aspiration pneumonia. Standardized differences were used to assess the balance between the head-and-neck cancer patients and non-cancer controls. The standardized differences for all covariates were less than 0.02 (data not shown) suggesting an acceptable match.

Kaplan-Meier survival plots and multivariate Cox proportional-hazard models were used to estimate the association of aspiration pneumonia with overall survival. The multivariate Cox-model for overall survival included all the covariates listed in Table 2. All statistical tests were two-sided, and p<0.05 was considered significant. Analyses were conducted with SAS version 9.4 (SAS Institute Inc, Cary, NC).

Results

Among the 3,513 head-and-neck cancer patients in this study, 801 patients developed aspiration pneumonia after radiation with a 5-month median time from initiating treatment to an aspiration pneumonia event. Among the 3,513 matched non-cancer controls, 272 developed aspiration pneumonia. As depicted in Figure 1, the 1-year and 5-year cumulative incidence of aspiration pneumonia was 15.8% (95% CI, 14.6–17.0%) and 23.8% (95% CI, 22.3–25.3%) for head-and-neck cancer patients and 3.6% (95% CI, 3.0–4.2%) and 8.7% (95% CI, 7.7–9.7%) for the non-cancer population, respectively.

Figure 1.

Cumulative incidence of aspiration pneumonia in patients with head-and-neck cancer (N=3,513) compared to non-cancer matched pairs (N=3,513).

Univariate analysis identified multiple risk factors for aspiration pneumonia of which select predictors are highlighted in cumulative incidence plots shown in Figure 2. Multivariate analysis (Table 3) revealed multiple independent risk factors for aspiration pneumonia including increased age, male gender, delivery of care at a teaching hospital, and tumor location in the nasopharynx or hypopharynx. Lower rates of aspiration pneumonia were identified for patients receiving treatment in the South (versus the West) and for patients with surgery prior to radiation. Of note, disease stage, type of chemotherapy, sequence of chemotherapy with radiation, or the delivery of IMRT did not impact the rate of aspiration pneumonia.

Figure 2.

Cumulative incidence of aspiration pneumonia among head-and-neck cancer patients by tumor site (A), Charlson comorbidity score (B), patient age at diagnosis (C) and care received at a teaching hospital (D).

Subsequent evaluation focused on the morbidity and mortality associated with aspiration pneumonia. Among the head-and-neck cancer patients who experienced aspiration pneumonia, 674 patients (84%) had events associated with hospitalizations with a 10 day median length of stay. Among those cancer patients hospitalized, 301 patients (45%) were admitted to an ICU during their hospitalization. Among cancer patients who developed aspiration pneumonia, the 30-day mortality rate after developing aspiration pneumonia was 32.5% (Figure 3A). Head-and-neck cancer patients who experienced aspiration pneumonia had significantly decreased overall survival compared to those without an aspiration event (Figure 3B). The 5-year survival from diagnosis among those with aspiration pneumonia was 15.2% compared to 32.3% among those without aspiration pneumonia (p < 0.001). Multivariate analysis controlling for potential confounders found that aspiration pneumonia after chemoradiation was associated with a 42% increase in the risk of death (HR=1.42, p < 0.001) compared to patients without aspiration pneumonia.

Figure 3.

Kaplan-Meier curves for overall survival. Figure 3A represents survival from the date of aspiration pneumonia among head-and-neck cancer patients who developed aspiration pneumonia (N=801). Figure 3B represents survival from the date of diagnosis among patients with head-and-neck cancer stratified by whether patients developed aspiration pneumonia (N=801) or not (N=2,712).

Discussion

This present US population-based study demonstrates substantial rates of aspiration pneumonia after concurrent chemotherapy with radiation in older patients with head-and-neck cancer. Previously reported rates of aspiration pneumonia for head-and-neck cancer patients after chemoradiotherapy have varied across study populations. A recent single-center study of 324 consecutively treated head-and-neck cancer patients from 2006 to 2008 reported that 5.3% patients developed aspiration pneumonia within in the first year after radiation with or without concurrent weekly cisplatin ¹². Eisbruch et al. reported 23% of their 29 patients treated with radiation and concurrent gemicitabine developed aspiration pneumonia within 14 months of completing their treatment [9] while Hunter et al. reported 15% of their 72 patients with locally advanced oropharyngeal cancer treated with IMRT and carboplatin/paclitaxel developed aspiration pneumonia within 24 months of treatment ¹¹. A larger Taiwanese retrospective study of 15,894 head-and-neck cancer patients based on insurance claims reported only a 5% incidence of pneumonia, most commonly associated with aspiration, within three months of treatment. ¹³ However, only 15% of the patients were older than 65. This current study found higher rates of aspiration pneumonia than the study from Taiwan, but identified similar risk factors such as age, male gender, and hypopharyngeal tumors. The range of incidence rates reported for aspiration pneumonia may be due to differences in patient characteristics, treatment specifics, follow-up, and definitions of aspiration pneumonia.

However, aspiration pneumonia is a relatively common event even among older adults without cancer. For elderly patients, the etiology of aspiration likely stems from multiple causes though data suggest that the progressive age-related loss of protective swallowing reflexes may be contributory ²⁷. Among patients with head-and-neck cancer, the etiology of aspiration pneumonia is likely multifactorial; however, radiation induced dysphagia presumably plays a central role ²⁸. Radiotherapy can produce fibrosis of the treated pharyngeal tissues resulting in nerve impairment and muscular injury causing generalized weakness and uncoordinated swallowing. Additionally, radiation induced xerostomia reduces salivary flow which could also impair the swallowing mechanism thus increasing the risk of aspiration pneumonia.

Radiation may also represent a unique risk factor for the development of aspiration pneumonia compared to other treatment modalities for head-and-neck cancer. A recent population-based study of 93,663 patients who underwent surgery for head-and-neck cancer found only a 2% rate of aspiration pneumonia ²⁹. Our study also found a lower risk of aspiration pneumonia for those treated with surgery prior to radiation though these findings could represent selection bias in that patients treated with surgery may have better overall health, less comorbidity, superior performance status, or more favorable cancer characteristics compared to those not receiving surgery. We also identified an increased risk of aspiration pneumonia in patients receiving care at a teaching hospital, which may reflect differences in un-measured patient characteristics. One could hypothesize that teaching hospitals may care for sicker or more complex patients which could explain the increased risk of aspiration in this sub-population.

Beyond the increased risk of aspiration with chemoradiotherapy, this study highlights the potential of aspiration pneumonia to substantially impact survival. This current study found significantly increased mortality after aspiration pneumonia. Other studies evaluating survival after aspiration pneumonia in non-cancer patients have found similar increases in the risk of death. A recent single institution retrospective study of 628 patients with aspiration pneumonia found a 30-day mortality rate of 21% though with a higher rate of hospitalization and ICU admission (99% and 38%, respectively)³⁰. Similarly, a multicenter retrospective cohort study of patients admitted with pneumonia found that aspiration pneumonia was the greatest risk factor for 30-day mortality with an adjusted hazard ratio of 5.69 ³¹.

The morbidity and potential impact on mortality of aspiration pneumonia after chemoradiotherapy raises the question of whether action can be taken to reduce the risk of this adverse event. Radiation techniques that spare the constrictor muscles involved with swallowing show potential to reduce dysphagia, a risk factor for aspiration pneumonia ^{32, 33}. Prospective studies to reduce the radiation dose in subsets of head-and-neck cancer patients are ongoing, and hopefully the decreased radiation dose will reduce toxicity without sacrificing local tumor control. Further research should focus on reducing the volume of tissue electively irradiated in head-and-neck cancer for patients with lower risk of regional tumor spread. Beyond modifying radiation fields, a different approach could concentrate on the preservation of swallowing function with early involvement of speech therapy and rehabilitation among patients at high risk of aspiration. Routine swallowing evaluations to screen for dysphagia among this patient population might also be recommended to select patients who are at risk for this complication. These interventions deserve further study as they stand to potentially improve survivorship in patients with head-and-neck cancer.

While this study characterizes the risk of aspiration pneumonia in a large cohort of head-and-neck cancer patients, there are limitations to the presented results. This study population only reported on patients over 66 years old and therefore, these findings cannot be generalized to a younger patient population. Medicare data is collected for administration purposes, and consequentially, confounding factors such as smoking status, dietary intake, body habitus, and performance status were not collected and could not be analyzed. Similarly, SEER-Medicare data does not include information regarding the doses of chemotherapy or radiotherapy delivered which could impact the rate of aspiration pneumonia. Furthermore, the higher aspiration pneumonia rates in this patient population could arise from local cancer progression as opposed to a radiation-induced toxicity. However, we found no correlation between tumor stage and aspiration rates. Additionally, this study could not directly assess the accuracy of the primary endpoint of aspiration pneumonia. While separate diagnosis codes exist for aspiration pneumonia and bacterial pneumonia, distinguishing these two diagnoses can pose a clinical challenge with misclassification potentially impacting our findings. Finally, SEER-Medicare data do not contain information on causality; therefore, future studies with additional longitudinal data are required to fully understand what drives the increased risk of aspiration in this cohort of patients.

Overall, this study represents the first US-based population study to define the incidence of aspiration pneumonia in patients with head-and-neck cancer treated with concurrent chemoradiotherapy contextualized with a comparison to the rate of aspiration in a matched sample of Medicare beneficiaries without cancer. This work also identified several risk factors that can be used to identify patients at a higher risk of developing aspiration pneumonia and suggested avenues for future mitigation of this clinically significant and potentially fatal complication of chemoradiation in head-and-neck cancer.