Search tips
Search criteria 


Logo of jathtrainLink to Publisher's site
J Athl Train. 2002 Apr-Jun; 37(2): 178–184.
PMCID: PMC164342

A 3-Dimensional Analysis of Face-Mask Removal Tools in Inducing Helmet Movement


Objective: To evaluate the performance of specific face-mask removal tools during football helmet face-mask retraction using 3-dimensional (3-D) video.

Design and Setting: Four different tools were used: the anvil pruner (AP), polyvinyl chloride pipe cutters (PVC), Face Mask (FM) Extractor (FME), and Trainer's Angel (TA). Subjects retracted a face mask once with each tool.

Subjects: Eleven certified athletic trainers served as subjects and were recruited from among local sports medicine professionals.

Measurements: We analyzed a sample of movement by 3-D techniques during the retraction process. Movement of the head in 3 planes and time to retract the face mask were also assessed. All results were analyzed with a simple repeated-measures one-way multivariate analysis of variance. An overall efficiency score was calculated for each tool.

Results: The AP allowed subjects to perform the face-mask removal task the fastest. Face mask removal with the AP was significantly faster than with the PVC and TA and significantly faster with the TA than the PVC. The PVC and AP created significantly more movement than the FME and TA when planes were combined. No significant differences were noted among tools for flexion-extension, rotation, or lateral flexion. The AP had an efficiency score of 14; FME, 15; TA, 18; and PVC, 35.

Conclusions: The subjects performed the face-mask removal task in the least amount of time with the AP. They completed the task with the least amount of combined movement using the FME. The AP and FME had nearly identical overall efficiency scores for movement and time.

Keywords: football equipment, cervical spine injury, axial load

Sports medicine health care professionals who are presented with a cervical spine injury (CSI) must manage the situation in a safe and effective manner. Should an athlete's airway be compromised at the time of injury, access to the airway is vital. Therefore, airway access must be achieved as quickly as possible while protecting the athlete from further injury. In sports such as football, a helmet and face mask are worn. This equipment, although designed to protect the athlete from injury, may inhibit the immediate care after an athlete sustains a head or neck injury by placing a physical barrier between the athlete and the athletic trainer or emergency medical technician (EMT). However, the cervical spine is taken out of neutral alignment when the football helmet is removed but the shoulder pads remain in place.14 Because it is important to obtain access to the athlete's airway should rescue breathing become necessary, it has become the practice of sports medicine professionals to remove the face mask from the helmet.318 Popular tools used for this purpose are polyvinyl chloride (PVC) pipe cutters, an anvil pruner, screwdrivers, EMT shears, rotary cutting devices, the Trainer's Angel (TA) (Trainer's Angel, Riverside, CA), and now the Face Mask (FM) Extractor (FME) (Sportsmedicine Concepts Inc, Geneseo, NY).19

Athletic governing bodies such as the National Collegiate Athletic Association (NCAA) and the National Athletic Trainers' Association (NATA) have developed and recommended certain guidelines for the care of a spine-injured athlete wearing a helmet.6,8,9 The NCAA Guidelines for Helmet Fitting and Removal in Athletics, which were developed in 1990 and revised in 1998, state that “proper on-the-field management of head and neck injuries is essential to minimize sequelae, expedite emergency measures and to prepare for emergency transportation.”9 The NCAA guidelines were revised with recommendations adapted from the Inter-Association Task Force for the Appropriate Care of the Spine-Injured Athlete in 1998. The guidelines explain that the helmet should never be removed in prehospital care unless one or more of the following 4 conditions exist:

  1. The helmet does not hold the head securely, such that immobilization of the helmet does not immobilize the head;
  2. The design of the sport helmet is such that even after removal of the face mask, the airway cannot be controlled or ventilation provided;
  3. After a reasonable period of time, the face mask cannot be removed; or
  4. The helmet prevents immobilization for transportation in an appropriate position.

The Inter-Association Task Force has recently published a document that explains in further detail the techniques and rationale behind proper care of the spine-injured athlete.6

The objective in removing a face mask is to create little or no movement of the head and neck while cutting through the face-mask loop straps in as short a time as possible. Removal of the face mask can be achieved through the use of a face-mask removal tool, such as a screwdriver or anvil pruner or devices specifically designed for the task, such as the TA.3,1222 While the FME appears to provide an effective method of face-mask removal compared with other previously studied tools, data evaluating its performance are limited.17,19

Effective CSI management begins with the recognition, initial care, and understanding of what is involved. The aspect of time is vital to the hope for neural recovery. Only by increasing the knowledge and understanding in the pathomechanics and pathophysiology of CSI will sports medicine practitioners be able to prevent further injury from occurring and be more prepared to manage an injury should it occur. Face-mask removal may make the situation worse if it is not done correctly. Athletic trainers and other sports medicine practitioners may be able to perform the face-mask removal technique more effectively when using a tool that has been scientifically shown to work.

A face-mask tool that allows for quick and safe removal of the face mask meets the needs of the athletic trainer, EMT, and any other qualified person. Athletic trainers and other sports medicine professionals must remember that extraordinary situations may present themselves at any time and, therefore, they must be prepared for any situation that may occur. Having the right tool to do the job is only the beginning. Proper preparation takes practice and planning.

The purpose of our study was to compare a variety of face-mask removal tools to determine if they differ in the time required to remove a face mask and in the amount of helmet movement that they produce.



Eleven certified athletic trainers, recruited from among the local population of sports medicine professionals, served as subjects for this study (Table (Table1).1). All subjects had at least a minimal degree of prior experience in face-mask removal. Subjects with any significant orthopaedic or neural pathologic condition of the upper extremities were excluded from the study, as were subjects who had extensive use and experience with any of the 4 tools used in this study. The latter was determined by asking each subject if he or she had prior experience with any tool in removing or retracting a face mask in a practice or real-life situation. Initial familiarization trials also helped to identify overqualified subjects. Standard institutional review board approval for this project was received through the Human Subjects Review Board at the University of Toledo. Each subject read and signed an informed consent form before participating in the study.

Table thumbnail
Table 1. Subject Demographic Means (SD)

Each of the 11 subjects performed the face-mask removal task with each of the 4 face-mask removal tools: anvil pruner (AP) (Scotts, Columbus, OH), FM Extractor (FME), PVC pipe cutters (PVC) (Anderson's Co, Maumee, OH), and the Trainer's Angel (TA) (Figure (Figure1).1). The subjects' performance in using each of the 4 tools to remove a face mask was analyzed. We used a random-order assignment of tasks by having each subject draw for the tools to be used in which order. Two groups of dependent variables were assessed: those associated with time and those associated with helmet movement.

Figure 1
Tools. Left, PVC pipe cutter. Bottom, Trainer's Angel. Right, Anvil pruners. Top, FM Extractor.


We used a 6-camera, EVa Hi-Res 3-D kinematic video analysis system (Motion Analysis Inc, Sacramento, CA) to collect the video data. Data-collection and processing software was the Motion Analysis Inc EVa, version 511. The duration of face-mask removal was measured using a hand-operated stopwatch. A standard football helmet (Riddell Co, Elyria, OH), shoulder pads (Douglas Protective Equipment, Houston, TX), and Schutt Sports (Batesville, MS) Armourguard face-mask loop straps were used in the testing.


The independent variable was the type of face-mask removal tool used, which consisted of 4 levels, the AP, FME, PVC, and TA groups. Two dependent variables were measured. Displacement of the head was measured in degrees of difference for the following planes: transverse (left-right) rotation of the head, bilateral (left-right) flexion, anterior-posterior (flexion-extension), and combined movement from all 3 planes. The total amount of time to complete the task was recorded in seconds.

Experimental Setup

All trials were performed in the University of Toledo Applied Biomechanics Laboratory. The football player (model) was placed on a thin indoor rug in the supine position on the floor in a designated area. Surrounding the model and subject were the 6 cameras in the video capture system. The positioning of the cameras resulted in a field of view that was approximately 2 m long, 1–1/2 m wide, and 1 m high. This size and shape of the capture volume was designed to ensure that the markers on the model were visible from all cameras. The model was oriented on the floor in such a way as to be aligned with the laboratory's coordinate system. Thus, the axes for the movements of flexion-extension, rotation, and lateral flexion of the head at the cervical spine were consistent with the X, Y, and Z axes of the laboratory.

The data-collection process for the 3-D system required establishing a specific project within the software for the task. Project setup included calibration of the area in which the task was performed, identification of the markers, and creation of links among those markers. The video marker set consisted of 4 markers attached to the helmet and 4 markers attached to the torso (Figure (Figure2).2). On the head, 2 marker wands were placed bilaterally on the front of the helmet, approximately 7.62 cm (3 in) from the midline of the forehead. Two other markers were placed along the midline of the helmet above the model's forehead, one attached to an extension placed approximately 12.7 cm (5 in) in front of the model's face. On the torso, markers were attached directly over the anterior aspect of the acromion processes by means of a wand that extended through a window cutout in the shoulder pads. Additionally, 2 markers were placed along the midline of the body of the sternum. This arrangement of markers allowed the 3-D system to track all movements of the head and torso and provided 3-D data in all planes of motion. This was accomplished by creating segments between the markers to represent the anterior-posterior head, medial-lateral head, shoulders, and sternum.

Figure 2
Anterior view of football player with markers applied.

Data Collection

Before video data-collection trials, a cube and wand calibration was performed in the area where the trials were to be executed. The first step after calibration involved collecting data on the static position of the model before any interaction with the test subject. This was done because each time the task was completed, the helmet needed to have the face mask reattached by replacement of the loop straps. It was impossible to perform this task for each trial without removing the helmet from the model, which subsequently resulted in movement of the head and torso. Therefore, by recording a brief static trial, the system had an exact position of the model's head to serve as a baseline before the subject trial.

During each subject trial, 3-D video data were recorded at 60 Hz and used to assess the amount of movement produced at the head caused by the tool cutting through the loop straps. Data collection was initiated 15 seconds into the start of the task and lasted for 30 seconds. The 3-D video collection was designed to capture only the movement data produced by the tools cutting through the loop straps. If the whole trial had been analyzed with 3-D video, tool performance would not have been based on the subject's ability but would have included movement due to other factors. The ideal time to collect video was determined to be 30 seconds after initial movements and before face-mask retraction efforts at the end. Movement is seen at the beginning of the trial as the subject “settles in.” Especially in the case of the PVC pipe cutters, the subjects often had to reratchet and adjust the tool several times before making an initial cut after the trial had already started. The later portions of the trial consisted of movement related to efforts to swing the face mask up. However, there was no clear-cut separation between pure cutting efforts and pure face-mask manipulation efforts. Subjects often put the tool down, attempted to retract the bar away from a loop strap, and picked the tool back up to make another cut. Therefore, the 3-D data are based on a sample of movement occurring during the face-mask retraction process when the tools were being used to cut through the loop straps.

Data Postprocessing

Each of the files containing the raw data from each trial was processed to create complete paths representative of the 3-D trajectories of each marker. These were smoothed with a Butterworth low-pass filter with a cut-off frequency of 3 Hz. Through the use of the segments created within the project of the EVa data capture system, angles of interest were defined and measured for displacement. Displacement from a neutral position was determined separately for each trial using the associated static trial position. For flexion and extension of the head, the software analyzed the relative position of the anterior-posterior head segment to the sternum segment. A neutral or starting position was approximately 0°. Any changes in the angle (positive or negative) between these 2 segments represented flexion (+) or extension (−). In order to define rotation of the head from left to right, the medial-lateral head position was compared with the shoulders. A positive increase from the neutral position represented rotation to the left. The neutral static position of the anterior-posterior head relative to the shoulders was approximately 0°. For lateral flexion, the anterior-posterior head segment was observed for its displacement relative to the shoulders. Any change in angle from a neutral position (approximately 90°) represented displacement in lateral flexion to the right or left.

Testing Protocol

Subjects reported to the Applied Biomechanics Laboratory of the University of Toledo on a separate day to practice the techniques and become familiar with the equipment and methods. The actual data collection was performed 2 to 4 days after the familiarization trial.

We gave exact, standard instructions to each of the subjects concerning the task procedures. Manufacturer's instructions for use of the FME and TA were also provided. The subject removing the face mask was required to cut through the 2 lateral face-mask loop straps and physically retract the face mask away from the face by rotating it on the anterior, or top, 2 loop straps. Subjects stabilized the head using their knees, and if able to, placed their free hand on the helmet while positioning the tool. Subjects were not allowed to move from their original position. As determined by a prerandom task selection (ie, the subject chose the order of trials randomly before performing the task), the subject retracted the face mask 4 times. Each retraction was performed with a different tool. Subject fatigue across trials was controlled by randomizing the order of tools used and by providing a rest period of 2 minutes between trials. We emphasized to the subjects, before they performed the trial, that their objectives were to create little or no movement of the head and to complete the task in as short a time as possible. We offered no encouragement or assistance during the trials. The model was instructed to avoid resisting movement. In addition, the model wore blackened goggles to keep him from seeing which tool was being used. Each trial began when the subject picked up the tool and ended when the subject rotated the face mask away from the 2 lateral loop straps. This indicated the beginning and end of each trial and served as the starting and ending point for timing the task. We maintained sharpness of the tools throughout the study with a standard tool sharpener based on manufacturers' recommendations.

Statistical Analysis

The 3-D and time data analysis consisted of a multivariate analysis of variance, set up as a simple 1-way analysis with repeated measures (probability level = .05) followed by a post hoc analysis. The 4 levels of the independent variable consisted of AP, FME, PVC, and TA groups. The dependent variables measured included (1) combined movement from all 3 planes, (2) maximum anterior flexion and posterior extension, (3) maximum right and left transverse-plane rotation, (4) maximum left and right lateral flexion, and (5) time to complete the task.

We performed statistical analyses on the SPSS 10.0 versions software for Windows (SPSS Inc, Chicago, IL) on IBM-compatible computers. Additionally, we calculated an efficiency score16 for all tools combining the movement and time variables.


Overall, our results indicated that the FME had the least amount of resultant head movement. However, none of the movement variables demonstrated statistical significance among tools. The combined movement variable was created by combining the resultant degrees of displacement from each of the 3 planes (Figure (Figure3).3). When planes were combined for each tool, no significant overall effect was found (F3,8 = 0.3647, P = .778). Movement at the head, as determined by displacement in the 3 separate planes, is shown in Figures Figures4,4, ,5,5, and and6.6. Means and standard deviations for the head-motion variables are presented in Table Table2.2. There was no overall significant effect for the flexion-extension (F3,8 = 1.19, P = .373), rotation (F3,8 = 3.425, P = .073), or lateral flexion (F3,8 = 2.88, P = .103) planes among tools. However, we found a significant effect when comparing each of the 3 planes with each other (P = .035), Post hoc analysis revealed that the most movement occurred in the lateral-flexion plane and was significantly greater than the rotation plane (P = .027).

Table thumbnail
Table 2. Mean (SD) Head Motion in Degrees by Tool (n = 11)
Figure 3
Flexion and extension movement of the helmet. AP indicates anvil pruner; FME, FM Extractor; PVC, PVC pipe cutters; TA, Trainer's Angel.
Figure 4
Rotation movement of the head to the left and right.
Figure 5
Lateral-flexion movement of the head to the left and right.
Figure 6
Combined head movement from within 3 planes.

Time to complete the task is shown in Figure Figure7.7. When using the AP (mean = 105.91 seconds), the subjects performed the face-mask removal task in the least amount of time, followed by the FME (mean = 176.82 seconds), TA (mean = 184.18 seconds), and PVC (mean = 256.55 seconds) (Figure (Figure7).7). The multivariate test revealed an overall significant effect for time (F3,8 = 12.007, P = .002). Time using the AP was significantly faster than with the PVC and TA, and time using the TA was significantly faster than with the PVC.

Figure 7
Average time to complete the face-mask retraction. AP indicates anvil pruner; FME, FM extractor; PVC, PVC pipe cutters; TA, Trainer's Angel.

Using an efficiency formula taken from Knox and Kleiner,16 we combined the movement and time variables to better represent the tool's overall performance. Combined movement was multiplied by time and divided by 10. Results (lower score = higher efficiency) revealed that the AP had an efficiency score of 14, followed by the FME (15), TA (18), and PVC (35).


A properly fitted football helmet should not allow for helmet movement without head movement. Motion at the head occurs due to movement in the cervical spine. Therefore, it is inferred that helmet movement results in head movement, which, in turn, requires movement in the cervical spine. However, our project did not directly measure cervical spine movement and, therefore, movements measured are strictly referred to as head movement.

Subjects were instructed to perform the retraction procedure in the same way while minimizing movement of the head and neck. The model wearing the equipment was instructed not to move throughout the testing. As long as these variables were controlled, the primary cause of movement was the ease or difficulty with which the loop straps were cut by the tools. However, being in the laboratory as opposed to on the actual playing surface with an injured athlete may have affected the subjects' performance, either positively or negatively.

The model wearing the football equipment was instructed not to move throughout the testing process. During actual CSI management, a medical professional would instruct the athlete not to move. Athletes in a CSI situation may not react the same as the model during the study. Thus, it is possible that they would not be as calm or willing to cooperate, considering the seriousness of the situation. External validity may be affected due to a lack of ability to generalize the laboratory situation to a real-life injury situation. In addition, we tested only certified athletic trainers, excluding any generalization of the results to student athletic trainers or other professionals.

Thirty seconds was determined to be an adequate amount of time to provide a representative sample of the movement during the tool's cutting component during the full task. Initially, it was our intent to analyze the entire task. However, if movement were assessed from the very beginning to the end of face-mask retraction, we would no longer be measuring only the tool's effectiveness but would have included initial and terminal movements not directly related to the tool's ability to cut through the loop straps. Therefore, the video-collection time was intended to capture the helmet movement produced by the tools cutting through the loop straps. Most of this movement occurred in the middle of the trial. Initial movement unrelated to the tool's cutting occurs in the beginning of the trial as the subject “settles in.” This is seen especially in the case of the PVC; the subjects often had to reratchet and adjust the tool several times after the trial had already started. The latter portions of the task consisted of movement related to efforts in swinging the face mask up and away from the loop straps. The ideal time to collect video data was determined to be around 30 seconds after initial movements and before face-mask retraction movements near the end. In effect, our 30-second window represents a sample of the movement that occurs during the task but is focused on capturing the movement created by the actual cutting of the loop straps. In addition, due to the length of time it typically took (2 to 3 minutes or more) to retract a face mask, the data files would have been much too large for storage and analysis.

Face-Mask Removal and Related Factors

Research on face-mask removal has primarily been performed by Kleiner.12,1618,20,2225 Knox and Kleiner16 rated the performance of several face-mask removal tools. By placing the head of the subject within a football helmet and on top of a forceplate, they were able to detect ground reaction forces placed on the head by the tools during the face-mask removal process. Knox and Kleiner16 attempted to quantify the movement placed on the head as a result of the use of the specific tools. Subjects included certified and student athletic trainers and EMTs. They rated the performance of the AP, the TA, a screwdriver, and a utility knife. The utility knife was removed from the study due to injuries to 2 practitioners during the process. The screwdriver produced the least amount of movement at the head, while the AP and TA created slightly more movement. In addition, the subjects rated their satisfaction with each tool and indicated their preference for the AP over the TA and screwdriver. No differences were found in the time or satisfaction with tools among any of the subject groups. However, it is interesting to note that the student athletic trainers performed the task in the least amount of time and with less resultant head movement than the certified athletic trainers.

When we conceived our study, no previous literature existed in any form for the assessment of the FME. However, 2 studies on the FME have recently been performed.17,19 Hoenshel et al19 investigated the performance of the FME compared with the AP and TA in time and satisfaction rating. They discovered that the FME was not superior to the AP in either time or rating but that subjects preferred to use the FME over the TA.

The design of the FME allows it to be used with varying methods; the manufacturer describes a primary technique and 2 alternative techniques.17 The primary technique is similar to other tools in having the cutting and noncutting surfaces aligned on either side of the loop strap between the face-mask bars. Another technique is to have the curved portion of the noncutting surface placed around the face-mask bar adjacent to the loop strap and cutting into the loop strap from the other side. The third technique involves placing the tool and each end on either side of the loop strap but on top of the face-mask bar. This method usually requires the athletic trainer to make more than one cut. Therefore, Angotti et al17 investigated the various techniques using 37 student athletic trainers. Each subject performed each task one time in a random order. The primary and second alternative methods differed significantly in time and satisfaction, with the primary method being neither the most preferred nor the most efficient.

Block et al18 evaluated tool performance with various tools and various types of loop straps. It is important to note that they found differences in tool performance with different loop straps. Future research of this type should include the use of various loop straps, as they may affect the use of the tool and alter the degree of helmet movement.

Head Movement

In our study, the FME created the least amount of movement in all planes, although the results were not statistically significant. Knox and Kleiner16 determined total movement of the head through deviation of the center of pressure of the helmet and head resting on the force platform. The TA produced significantly more movement of the center of pressure than the screwdriver and AP. Their results are not consistent with our results. It is difficult to relate the results from both studies, considering the differences in methods and instrumentation. This investigation, which was designed to quantify the movement at the head during face-mask retraction, is unique in its methods and instrumentation. To our knowledge, no prior research has adopted 3-D videography to assess the amount of movement associated with the use of face-mask removal tools. Knox and Kleiner16 investigated head movement during face-mask removal but through the use of a forceplate.

Knox and Kleiner16 assessed the efficiency of the TA, AP, and a screwdriver for head movement. Although they found the screwdriver to have the least amount of movement, we did not include it in our tool selection because the screwdriver may be unreliable for the task of face-mask removal should the T-nut and screw assembly in the helmet become faulty.16

The FME appeared to produce the least amount of average displacement for flexion and extension. However, there were no significant differences among the tools in this plane. Typically, the line of force being applied to the helmet by the tool is in a lateral-to-medial direction, so there is a diminished tendency for movement in the anterior-posterior plane. However, Knox and Kleiner16 found the TA to be significantly greater in inducing head movement than the screwdriver in flexion and extension. Again, the results from the 2 studies are not comparable due to the type of instrumentation used for quantifying head movement and, therefore, the results are difficult to relate.

The FME produced the least amount of movement in rotation and lateral flexion. Knox and Kleiner16 found no significant differences in rotation among the screwdriver, TA, and AP. They were not able to define lateral-flexion movement. In the rotation and lateral-flexion planes of movement, the researchers noted a tendency for left rotation and left lateral flexion as opposed to right rotation and right lateral flexion. This was evident in the overall total displacement angles, which increased from the neutral position. For example, left rotation was represented by an angle above 0°, and right rotation was less than 0°. Left lateral flexion was represented by an angle greater then 90° and right lateral flexion by an angle less than 90°. Almost all of the trials evidenced rotation and lateral flexion to the left. This tendency was due to the fact that all but one of the subjects was right-hand dominant. Thus, most subjects approached the lower right loop strap first during the task. A force applied from the right in the medial-lateral plane results in movement to the left. Since data collection began after 15 seconds from the start of the task, most of the force being applied during that 30 seconds was coming from the right side of the helmet for those trials that took a longer amount of time.

Additionally, an important finding for health care professionals is that the lateral-flexion plane appears to be where the most movement occurred during the task. This should increase our awareness of the susceptibility for movement in this plane.

Although each tool induced relatively little movement, sometimes less than 1°, it is still important to realize that this may be all that is necessary to damage the spinal cord. We analyzed the data from the 3-D analysis for degrees of difference between the maximum and minimum measured angles during the 30-second trials. The actual difference between the maximum and minimum angles was calculated, totaled, and averaged across subjects for each tool. The amount of total displacement in that observed amount of time was usually less than 10°. However, when considering the space within the spinal canal and the relative sensitivity and delicacy of the spinal cord, very little movement is needed to compromise it. This becomes truer when there is already decreased space in the spinal canal due to inflammation or a loss of structural integrity.


The findings revealed that the subjects using the AP performed the face-mask removal task the fastest, followed by the FME. Those using the AP were significantly faster than those using the PVC and TA, and those using the TA were significantly faster than those using the PVC.

Much of the previous research into face-mask removal and retraction focused on the amount of time it takes to complete the task.1317,20,23,24 Those studies included tools such as the AP, TA, PVC, EMT shears, scalpels, screwdrivers, and utility knives. Discounting the screwdriver for the reasons explained previously, our findings are consistent with other research in that those subjects using the AP tool performed the task the fastest. However, only one study has included an evaluation of the FME.19 In that study, use of the FME was second in speed to the AP but faster than the TA. Our results are consistent with those results.

Knox and Kleiner16 found the AP to induce the least amount of movement, second to the screwdriver, when compared with the TA and PVC. However, our results indicate that the AP produced significantly more movement in the combined variable when compared with the FME. Conversely, the AP performed better than all of the other tools for the time variable. It is possible, however, that because the AP was faster than the other tools, there may have been some trials in which the AP task was nearly completed and the subject was attempting to remove the face mask. Therefore, we analyzed the first 15 seconds of the 30-second video trial for all tools in order to determine whether the AP data were affected, and the results were the same for the amount of movement produced by each tool.

This brings up another issue: which is more important, speed of retraction or movement created at the head during retraction? Our research does not attempt to answer this question and treats each variable equally using the efficiency score. However, this issue is important, and research in this area should address both variables when assessing tool performance.

Knox and Kleiner placed a notch in the noncutting edge of the AP.16 However, this method is subjective in its process, and many types of AP tools are available. Therefore, since no standard form of altering a tool such as the AP has been developed, the tools used in this study were left in their original form.

In an actual CSI situation, the athletic trainer or other personnel may have others to help in the management process. A second person could aid in stabilizing the head and neck while the first person performs face-mask removal. However, athletic trainers should be prepared for the worst-case scenario. A worst-case scenario in CSI is being the only person, or the only qualified person, in the immediate area to treat the athlete. Furthermore, it is difficult to control for confounding factors during data collection that may be introduced by having a second person applying pressure on the helmet. Additionally, the 3-D data capture volume in which the task was being performed needed to be free from other objects that could block the view of cameras.

Being able to perform such an important skill would logically require that those performing face-mask removal practice the skill in order to be more effective. Times assessed during the familiarization trial and even during data collection often exceeded 2 and 3 minutes. This is unacceptable in an emergency situation. Kleiner et al13 looked at the effects of practice on face-mask removal skills, specifically investigating the effects of practice on time. This study had 22 student athletic trainers serve as subjects. The subjects used both a TA and AP and were tested for time before and after 7 days of practice. Ratings of satisfaction were also analyzed after each trial. Posttest means for both conditions of satisfaction and time improved significantly for each tool. Differences among the tools were also assessed for time and satisfaction and revealed that the AP was preferred and performed the task in the least amount of time. An interesting progression to this current investigation would be to analyze movement before and after a practice protocol.


Based on our results, we conclude that the AP performed the task in the least amount of time and the FME induced the least amount of head movement. The AP and FME were nearly identical in their overall efficiency scores.

This investigation can serve as a model for future research on face-mask removal and CSI management. Three-dimensional analysis is an effective tool for detecting minuscule changes in movement. Thus, the use of 3-D analysis can provide valuable feedback for face-mask removal and CSI management techniques.

Additional research is necessary on this critical skill and should remain current with the developing advances and trends in face-mask removal guidelines and tool design. Future investigators should include a higher number of subjects representing other populations (EMTs, athletic training students) and incorporate other types of loop straps.

We highly recommend that a skill as vital and important to CSI management as face-mask removal be incorporated into the practical section of the NATA Board of Certification Examination.


We thank Dr Robert Kenefick, Department of Kinesiology, University of New Hampshire, for his assistance with the statistical analyses.


  • Donaldson W F, 3rd, Lauerman W C, Heil B, Blanc R, Swenson T. Helmet and shoulder pad removal from a player with suspected cervical spine injury: a cadaveric model. Spine. 1998;23:1729–1733. [PubMed]
  • Gastel J A, Palumbo M A, Hulstyn M J, Fadale P D, Lucas P. Emergency removal of football equipment: a cadaveric cervical spine injury model. Ann Emerg Med. 1998;32:411–417. [PubMed]
  • Palumbo M A, Hulstyn M J, Fadale P D, O'Brien T, Shall L. The effect of protective football equipment on the alignment of the injured cervical spine: radiographic analysis in a cadaveric model. Am J Sports Med. 1996;24:446–453. [PubMed]
  • Swenson T M, Lauerman W C, Blanc R O, Donaldson WF, 3rd, Fu F H. Cervical spine alignment in the immobilized football player: radiographic analysis before and after helmet removal. Am J Sports Med. 1997;25:226–230. [PubMed]
  • Feld F, Blanc R. Immobilizing the spine injured football player. J Emerg Med Serv. 1987;12:38–40.
  • Prehospital Care of the Spine-Injured Athlete. National Athletic Trainers' Association; Dallas, TX: 2001. Inter-Association Task Force for the Appropriate Care of the Spine-Injured Athlete.
  • Maroon J C, Bailes J E. Athletes with cervical spine injury. Spine. 1996;21:2294–2299. [PubMed]
  • National Athletic Trainers' Association . Athletic Helmet Removal Guidelines. National Athletic Trainers' Association; Dallas, TX: 1998.
  • National Collegiate Athletic Association . Guideline 4F: Guidelines for Helmet Fitting and Removal in Athletics. National Collegiate Athletic Association; Indianapolis, IN: 2001.
  • Patel M N, Rund D A. Emergency removal of football helmets. Physician Sportsmed. 1994;22(9):57–59.
  • Kleiner D M. Football helmet face mask removal. Athl Ther Today. Jan, 1996. pp. 11–13.
  • Kleiner D M. Face mask removal vs. face mask retraction [abstract] J Athl Train. 1996;31:S32.
  • Kleiner D M, Almquist J L, Hoenshel R W, Angotti D D. The effects of practice on face mask removal skills [abstract] J Athl Train. 2000;35:S60.
  • Kleiner D M, Knox K E. An evaluation of the techniques used by athletic trainers when removing a face mask with the Trainer's Angel [abstract] J Athl Train. 1995;30:S7.
  • Knox K E, Kleiner D M. The effectiveness of EMT shears for face mask removal [abstract] J Athl Train. 1996;31:S17.
  • Knox K E, Kleiner D M. The efficiency of tools used to retract a football helmet face mask. J Athl Train. 1997;32:211–215. [PMC free article] [PubMed]
  • Angotti D D, Hoenshel R W, Kleiner D M. The most efficient technique for using the FM Extractor [abstract] J Athl Train. 2000;35:S60.
  • Block J J, Kleiner D M, Knox K E. Football helmet face mask removal with various tools and straps [abstract] J Athl Train. 1996;31:S11.
  • Hoenshel R W, Angotti D D, Kleiner D M. An objective evaluation of the FM Extractor [abstract] J Athl Train. 2000;35:S61.
  • Knox K E, Kleiner D M, McCaw S T, Ryan M. The effects of qualifications on the efficiency of football helmet face mask removal with various tools [abstract] J Athl Train. 1995;30:S7.
  • Rehberg R S. Face mask removal tools come in all shapes, styles. NATA News. Aug, 1999. pp. 8–9.
  • Almquist J L, Rehberg R S, Kleiner D M. An assessment of the response time prior to face mask removal [abstract] J Athl Train. 2000;35:S61.
  • Redden W W, Kleiner D M, Holcomb W R. The effects of gender, hand size and grip strength on the ability to remove a football helmet face mask [abstract] J Athl Train. 1998;33:S43.
  • Kleiner D M, Greenwood L D. The influence of hand size and grip strength on the ability to remove a football helmet face mask [abstract] J Athl Train. 1997;32:S50.
  • Kleiner D M, Sonnenberg R J. The influence of temperature on the ability to cut the football helmet face mask loop-strap attachment [abstract] J Athl Train. 1999;34:S84.

Articles from Journal of Athletic Training are provided here courtesy of National Athletic Trainers Association