Data from: Effects of chemoradiation and tongue exercise on swallow biomechanics and bolus kinematics
Data files
Sep 25, 2024 version files 2.77 MB
-
ChemRad_Breathing_Analysis.xlsx
9.53 KB
-
ChemRad_Jaw_Averages.xlsx
14.08 KB
-
ChemRad_Master_Sheet_Final_Organized.xlsx
10.92 KB
-
Fluoro_Protocol.docx
2.35 MB
-
Fluoro_Template_Copy.xlsx
19.98 KB
-
MorphoJ_Analysis_Protocol.docx
332.10 KB
-
README.md
7.91 KB
-
Respiration_Blinding_02092022.xlsx
25.45 KB
Abstract
Background: Common treatments for head and neck cancer (radiation and chemotherapy) can lead to dysphagia; tongue exercise is a common intervention. This study aimed to assess swallow biomechanics and bolus kinematics using a well-established rat model of radiation or chemoradiation treatment to the tongue base, with or without tongue exercise intervention.
Methods: Pre- and post-treatment videofluoroscopy was conducted on thirty-two male Sprague-Dawley rats treated with radiation/chemoradiation and exercise/no exercise. Rats in the exercise groups completed a progressive resistance tongue training paradigm. Swallow biomechanics, bolus kinematics, jaw opening, and post-swallow respiration were assessed.
Results: Both treatments impacted outcome measures; the addition of exercise intervention showed benefit for some measures, particularly in rats treated with radiation, vs. chemoradiation.
Conclusions: Radiation and chemoradiation can significantly affect aspects of deglutition; combined treatment may result in worse outcomes. Tongue exercise intervention can mitigate deficits; more intensive intervention may be warranted in proportion to combined treatment.
README: Updated Effects of Chemoradiation and Tongue Exercise on Swallow Biomechanics and Bolus Kinematics: Open Source Data
Description_of_the_data,_software,_and_file_descriptions
We have submitted our swallow and respiration patterns for the animals (ChemRad_Breathing_Analysis.xlsx); rat jaw aperture (opening) data (ChemRad_Jaw_Averages.xlsx); final organized data for bolus area, bolus speed, and mastication rate for pre and post WHAT? (ChemRad_Master_Sheet_Final_Organized.xlsx); bolus speed, bolus area, and mastication for each animal (Fluro_Template_Copy.xlsx); analysis of swallow-respiratory patterns of the animals in which the rater is blinded to timepoint and group (Respiration_Binding_02092022.xlsx); images showing the process we used to analyze the fluorscopic data (Fluoro_Protocol_Copy.docx); a written walkthrough of how we used MorphoJ to use CASM-R (MorphoJ_Analysis_Protocol.docx); statistical breakdown of collected data (PB_VELOC_GLM_Glen stats.docx); Processing and analyzing software ImageJ; and software used to analyze swallow biomechanics.
Abbreviations
HNC - Head and Neck Cancer
RAD - Radiation Associated Dysphagia
LRAD - Late Radiation Associated Dysphagia
VFSS - Videofluoroscopic Swallow Study
RT - Radiotherapy
CCRT - Concurrent Chemotherapy and Radiation Treatment
TE - Tongue Exercise
CASM-R - Computational Analysis of Swallow Mechanics adapted for the Rodent model
Software_Descriptions
1)Image J
This program was used to process and analyze video-recorded swallows (using java) both pre- and post-treatment and intervention for all animals included in the study to obtain the following outcomes: bolus speed, bolus area, mastication rate, jaw aperture, and post-swallow respiration.
2)MorphoJ
This program was used to analyze swallow biomechanics via the video-recorded swallows identified for analysis and converted in Matlab for morphometric analysis.
File_Descriptions
1)ChemRad_Breathing_Analysis.xlsx
ID: Randomized animal identification number
Treatment: Experimental condition describing what treatment the animal underwent.
CRT = Chemo therapy and radiotherapy
RT = Radiotherapy
Condition: Experimental condition denoting if the animal underwent tongue exercise or not.
Ex: = Exercise
NonEx = Non-Exercise
Pre_(E_or_I): A value measured before the animal underwent treatment
E = Exhale
I = Inhale
Post_(E_or_I): A value measured after the animal underwent treatment
E = Exhale
I = Inhale
NA = Diaphragm could not be visualized
2)ChemRad_Jaw_Averages.xlsx
ID: Randomized animal identification number
Jaw_Aperature_with_Calibration_Pre: Pre-treatment measurement of jaw opening
Jaw_Aperature_with_Calibration_Post: Post-treatment measurement of jaw opening
3)ChemRad_Master_Sheet_Final_Organized.xlsx
ID: Randomized animal identification number
Bolus_Area(Pre): The size of the swallowed amount post treatment
Bolus_Area(Post): The size of the swallowed amount measured post treatment
Bolus_Velocity(Pre): The velocity of the swallowed bolus measured before the animal underwent treatment
Bolus_Velocity(Post): The velocity of the swallowed bolus measured after the animal underwent treatment
Mastication_Rate(Pre): The amount of chews before the bolus was swallowed before the animal underwent treatment
Mastication_Rate(Post): The amount of chews before the bolus was swallowed after the animal underwent treatment
4)Fluro_Template_Copy.xlsx
Animal_ID: Randomized animal identification number
Swallow#: swallow 1-3 included for analysis
Frame_Start: The video frame number corresponding with the start of the animals swallow
Frame_End: The video frame number corresponding with the end of the animals swallow
Stable_Marker_Start: stable head marker used to account for movement during the swallow, prior to swallow initiation
Initation: intial movement of bolus/initiation of swallow
Stable_Marker_End: stable head marker used to account for movement during the swallow, post-swallow cessation
Point_Number: point number on frame during bolus tracking
Area: The area of the swallowed bolus
Mean: The mean value of
Min: The minimum value of
Max:The maximum value of
X: the value on the X axis of
Y: The value on the Y axis of
Slice: # of frames included in swallow (i.e. 1-6)
Correction: documenting if animal moved forward or backward during swallow - calculations adjusted accordingly per template
Point Number: point number re: bolus tracking (i.e. 1-6) as you click through frames of swallow
Initiation: frame and all related measurements from ImageJ when swallow was initiated
Multi-point_tool_number: same as point number above
X-axis: measurement provided from ImageJ "measure" function (control+m)
Y-axis: measurement provided from ImageJ "measure" function (control+m)
Distance(Pixels): distance in pixels re: bolus measurement
Distance(mm): distance in mm re: bolus measurement
Frames: video frames containing the current swallow being analyzed
Location: measure provided by ImageJ
Velocity: bolus speed
Calibration_Value: calibration value determined by dime measurement at beginning of video recording used in calculation for outcomes
All: overall measure of bolus speed
Average: average speed across frames
Moved_Forward: head moved forward between swallow initiation and end of swallow
Moved_Backward: head moved backward between swallow initiation and end of swallow
5)Respiration_Blinding_02092022.xlsx
ID:Randomized animal identification number
Frames_for_Swallow_1: The total number of frames it took for the animal to complete one full swallow
Pre_Swallow1_BlindCode: The code used in the blind evaluation process to replace the animals ID number, used to identify the corresponding video file of the animals swallow pre treatment
Frames_for_Swallow_2: The total number of frames it took for the animal to complete a second full swallow
Pre_Swallow2_BlindCode: The code used in the blind evaluation process to replace the animals ID number, used to identify the corresponding video file of the animals swallow pre treatment
Frames_for_Swallow_3: The total number of frames it took for the animal to complete a third full swallow
Pre_Swallow3_BlindCode: The code used in the blind evaluation process to replace the animals ID number, used to identify the corresponding video file of the animals swallow pre treatment
Post_Swallow1_BlindCode: The code used in the blind evaluation process to replace the animals ID number, used to identify the corresponding video file of the animals swallow post treatment
Post_Swallow2_BlindCode: The code used in the blind evaluation process to replace the animals ID number, used to identify the corresponding video file of the animals swallow post treatment
Post_Swallow3_BlindCode: The code used in the blind evaluation process to replace the animals ID number, used to identify the corresponding video file of the animals swallow post treatment
6)Fluoro_Protocol_Copy.docx
Word file (.docx) containing images showing the process of analyzing fluoroscopic data.
7)MorphoJ_Analysis_Protocol.docx
Word file (.docx) containing a written walkthrough of using MorphoJ to conduct CASM-R.
8)PB_VELOC_GLM_Glen_stats.docx
Word file (.docx) containing the statistical breakdown of the collected data.
9)Respiration_Glen_Stats.docx
Word file (.docx) containing the statistical breakdown of the animal's respiration data.
Methods
Thirty-two male Sprague-Dawley rats were randomized into four groups: concurrent chemoradiotherapy without exercise (CCRT), radiotherapy without exercise (RT), CCRT+ tongue exercise (CCRT+TE), and radiotherapy (RT+TE), with 8 animals per group based on a priori power calculations). All rats underwent videofluoroscopic swallow studies (VFSS) at baseline (prior to treatment administration/exercise intervention) and 12 weeks post-treatment +/- exercise intervention, thus serving as their own controls. Treatment consisted of RT delivered to an 8X12 mm2 area of the tongue base over the course of ten days using methods previously described (Xstrahl Small Animal Radiation Research Platform (SARRP) Irradiator). Administration of radiotherapy to this area targets a structure crucial for swallowing and allows exploration of how treatment may impact swallow function.
Each rat received 4.5 Gy/day, totaling 45 Gy – thus the biological effective dose was approximately 112 Gy. This fraction size was calculated using methods previously described, and allowed for maintenance of clinical relevance in the protocol of treatment delivery. In addition, rats in the CCRT groups received two doses of cisplatin (3mg/kg) in addition to RT – one dose prior to the first fraction of RT, and another prior to the sixth fraction. Cisplatin is a common chemotherapy drug and is often considered the standard of care in this patient population.49,50 Rats in the TE groups underwent an 8-week progressive resistance tongue training paradigm following treatment administration and a recovery period. Baseline and final VFSS recordings were used for data analysis via ImageJ. Rats were pair-housed in standard polycarbonate cages in the vivarium on a reversed 12:12 light/dark cycle. All animal work was done in accordance with NIH and university animal-use guidelines and approved by the University of Wisconsin School of Medicine and Public Health Animal Care and Use Committee (IACUC Protocol M005828).
Progressive Resistance Tongue Training Paradigm:
The progressive resistance tongue training paradigm is a well-established tool, analogous to lingual strengthening exercises employed in the clinical setting, in translational models such as aging, Parkinson disease, and stroke. Intervention involves training the rat to press a disc connected to a force transducer with a specified amount of force to obtain a water reward (Figure 1). To avoid adding undue stress during treatment administration, rats were not water restricted until after treatment. Following a 1-week (7 days) gradual water restriction, rats were acclimated to the apparatus and underwent an 8-week training paradigm. Maximum force (g) was calculated at baseline, and individualized force requirements were initiated at 20% and increased by 20% every two weeks, ending with 80% maximum force required at eight weeks. Rats were water restricted over the training period to motivate their participation in the exercise, and given access to water ad libitum for 3 hours/day, 7 days/week.
Videofluoroscopy:
Videofluoroscopic swallow study (VFSS) allows for radiographic imagining of bolus movement and swallowing physiology from the oral cavity to the esophagus. Quantitative measures of deglutition including bolus speed, bolus area, and mastication rate have been validated in prior animal studies as clinically-relevant measures of oral phase efficiency. Such outcomes provide information on potential physiological changes, thus enabling translation to human studies given anatomical similarities present between this model and humans. Further, assessment of swallow biomechanics and post-swallow respiration using VFSS recordings provide critical information regarding deglutition. Analysis of these outcomes can provide clinically-relevant insight into aspects of swallow physiology in both the oral and pharyngeal phases of swallowing.
All rats underwent baseline and post-treatment/intervention recordings to compare potential changes both within and between groups. Rats were food restricted prior to recording and acclimated to smooth peanut butter placed on the side of the home cage prior to recording day. To conduct VFSS, rats were individually placed in their home cage on an L-shaped platform perpendicular to the C-arm fluoroscope (Genoray ZEN-7000 C-Arm 3rd Gen Fluoroscope; Origin Republic of Korea, Vendor Mulit Inc., Ontario, CA). A mixture of peanut butter and barium was placed on a stage affixed to the side of the home cage; rats were allowed to consume the mixture ad libitum for approximately five minutes (Figure 2). X-ray was only initiated when the rat began to consume the mixture, as to avoid unnecessary radiation exposure. Swallows were recorded at 30 frames/second and analyzed offline, using MorphoJ for swallowing biomechanics, and ImageJ for all other measures.
Computational Analysis of Swallow Mechanics adapted for the Rodent Model (CASM-R):
Computational Analysis of Swallowing Mechanics (CASM) is a morphometric analysis software that has been used for studying swallowing biomechanics in humans with dysphagia, and has been adapted for use in rodents (CASM-R). CASM-R tracks markers at skeletal levers formed by the hard palate, 1st and 4th cervical vertebrae, and mandible (Figure 3A) at each frame, in order to describe body and jaw movements during the swallow. Skeletal lever tracking also serves to re-axis / “anchor” the position of soft tissue structures relative to the skeletal landmarks, allowing for more specific analysis of the isolated movement of swallowing structures, including the upper esophageal sphincter (UES) and base of tongue (BOT), without interference of gross body/jaw movements.
Assessment of swallow biomechanics can provide information on potential musculoskeletal dysfunction through analysis of key muscular and skeletal landmarks affiliated with swallowing, thereby guiding investigation of the underlying mechanisms treatment may have on swallow physiology. An initial group of sixteen male Sprague-Dawley rats (n = 4/group) were included for a preliminary pilot analysis, following methods previously described in Kletzien et al., 2017 and in alignment with published clinical work supporting significant outcomes with limited sample size, to determine the impacts of treatment and exercise intervention on swallow biomechanics. Briefly, three VFSS recorded swallows were selected for analysis. Eight muscular and skeletal landmarks were identified – the hard palate, C1, C4, the UES, three mandibular points (anterior, mid, and posterior), and tongue base – and mapped in each frame of the individual swallow for multivariate morphometric analysis. The pharyngeal phase was defined as the frame immediately preceding swallow onset until the bolus head made contact with the UES. Swallows in which the rat’s positioning was suboptimal for data analysis (i.e., obliqued, extraneous movement, or upper extremities blocking structures) were identified as outliers by MorphoJ and were subsequently excluded from final analysis, as inadequate positioning could provide incorrect data related to the positioning and movement of the key anatomical landmarks. In these cases, said excluded swallow was replaced with another from the same rat, selected at random. Following this preliminary pilot multivariate morphometric analysis, analysis of bolus kinematics and bolus area was completed to study the dynamic interaction on swallow function in a larger sample size. Given interpretation of CASM-R findings is limited, the subsequent inclusion of kinematic analysis allowed for more robust assessment and subsequent conclusions of the relationship between HNC treatment, tongue exercise intervention, and physiologic outcomes.
Bolus Kinematics and Bolus Area:
Prior to analysis, three swallows were selected per rat. Selection was based on the following criteria: unobstructed sagittal view, adequate positioning during recording (ideally little to no movement, rat positioned straight without curvature of body/movement of limbs), and visualization of bolus from oral cavity (mastication) through esophagus. To measure bolus speed (mm/sec), a marker was placed at the leading edge of the bolus (referred to as the head of the bolus), beginning at the frame immediately preceding swallow onset, and at each subsequent frame until the head of the bolus made contact with the UES. A marker was placed on the skull (rostral and posterior to the eye) at the frame before swallow onset, and again at the final frame of the swallow, in order to account for whole body movement. Bolus area was measured at the final frame of the swallow (when bolus head first encounters UES) as the slowed transit caused by impedance of the UES permits greater visualization of the bolus, with less interference of movement artifact on bolus edges. To further reduce measurement variability, the same bolus frame was measured 3 times, with the average used for final analysis. Pixel area was converted to mm2 using a calibration factor derived from a 17.9cm metal disc.
Mastication rate (cycles/sec) was calculated from the frame of minimum jaw opening (jaw closure) following bolus acquisition, through 5 complete cycles of jaw opening-closing, until the closure of the 5th cycle. All videos were analyzed by rater 1 (NSH), and a random subset of videos (10%) was analyzed by a second rater (AFP) to calculate inter-rater reliability. Agreement amongst the raters was excellent (ICC=0.99).
Jaw Opening:
Following results obtained from the assessment of mastication rate, the length of jaw opening (aperture) was measured to investigate the effects of treatment and intervention on the mandibular range of motion. 3 mastication cycles were selected, analyzed, and averaged per rat. The point of maximum jaw opening in a cycle was identified, and a line was drawn from the lowest point of the mandible (mid-anterior) to the highest point of the hard palate. Measurements were taken pre-and post-treatment/tongue exercise intervention for all groups.
Respiratory-Swallow Coordination:
The typical process of deglutition involves a post-swallow exhalation immediately following the swallow apnea period. HNC treatments can impact the coordination of breathing and swallowing, resulting in an unsafe post-swallow inhale that elevates the risk of aspiration. A previous study in our lab confirmed the presence of this aberrant post-swallow inhale pattern in a rat model of combined chemoradiation treatment, using a validated diaphragm tracking measure as a proxy for respiration. In the present study, an abbreviated version of this assay was used to determine the differential effects of CCRT, RT, and tongue exercise on respiratory-swallow coordination. The position of the diaphragm was visually tracked during the breathing cycle associated with each analyzed swallow, and a binary code was assigned for statistical analysis: post-swallow exhale (0; Ex) or post-swallow inhale (1; In). Raters, blinded to rat ID, group assignment, and time point, identified the onset frame and directionality of each phase: pre-swallow respiration (In/Ex), swallow apnea (cessation of movement), and post-swallow respiration (In/Ex). All files were analyzed by NSH, and 10% of all videos were analyzed by a second rater (LMR) for inter-rater reliability (IRR = 90%).
STATISTICAL ANALYSIS
Swallow Biomechanics:
Procrustes fit was generated to evaluate the distribution of shape change among rats and highlight potential outliers. Following identification and removal of outliers, a canonical variate analysis was conducted to determine biomechanical differences between the groups. A discriminant function post hoc analysis was completed to visualize said changes through corrected eigenvectors. Mahalanobis distance (D) was used to indicate significance. To account for multiple comparisons across 8 coordinates, a post hoc Bonferroni correction was performed with a corresponding set critical value of α<0.006.
Bolus Kinematics, Bolus Area, Jaw Opening, and Post-Swallow Respiration:
To compare the effects of treatment and tongue exercise on bolus speed, bolus area, mastication rate, and jaw aperture within (pre vs. post time points) and between groups, a two-way repeated measures analysis of variance (ANOVA; assumptions met) with pairwise comparisons and post hoc Bonferroni correction was conducted using SPSS (IBM, Chicago, IL). To assess potential changes in post-swallow respiration patterns, descriptive statistics were generated (frequency and percentage of post-swallow inhales across 3 trials per rat) to compare the two time points and four groups – RT without exercise, RT+TE, CCRT without exercise, and CCRT+TE. Given there were no aberrant post-swallow inhales during the pre-treatment and pre-intervention time point (as expected for healthy controls), McNemar’s test to assess repeated measures of a binary outcome could not be completed. As such, Fisher’s Exact Test was conducted to assess the differences between groups in the post-treatment and intervention time point. Further, to determine how bolus area and post-swallow respiration impacted post-bolus speed, an ANCOVA was conducted on post-bolus speed data, with the covariates being pre-treatment bolus speed, post-treatment bolus area, and post-swallow respiration using SAS. A critical value of α=0.05 indicated significance.