Authors: Damjan Slabe, Eva Dolenc Šparovec, Miha Fošnarič
Categories: Experimental Paper, Basic life support, Force, Alternative methods
Source: Resuscitation Plus
Authors: Damjan Slabe, Eva Dolenc Šparovec, Miha Fošnarič
•The method of compression affects both the magnitude of the force exerted and its angle relative to the vertical.•The maximal force during FCCs is ≈5 % lower than during HCCs, while the angle from the vertical is nearly twice as large.•FCCs are associated with significantly lower quality metrics for chest compressions.•The two-handed compression method is more effective in achieving adequate compression depth, rate, and vertical force.•On average, both methods produced compression parameters within or near the recommended guideline ranges.
Using the foot to perform chest compressions (CCs) as an alternative to using the hands was listed as an acceptable method in the 2005 American Heart Association (AHA) Guidelines.^1^ Foot chest compressions (FCCs), performed with the heel or ball of the foot, are recognised as useful in situations where a lone rescuer is unable to apply effective hand CCs, typically due to physical limitations^2^ or fatigue.^3^ However, FCCs have several disadvantages compared to the hand technique and may not be safely performed in certain environments. Concerns include balance, potential injury, and reduced effectiveness.2, 4, 5, 6 The distribution of force during CCs is related to compression quality and may influence the risk of injury.2, 6
With the evolution of resuscitation guidelines, the importance of CCs has increased.^7^ Current cardiopulmonary resuscitation (CPR) instructions recommend positioning oneself vertically above the adult victim’s chest and pressing down on the sternum to a depth of at least 5 cm, but not more than 6 cm.^7^ The effectiveness of CCs depends on correct hand placement, compression depth and rate, and full chest recoil. A recent study^8^ found that, under standardised conditions and over a realistic time span, the ‘leg-heel’ approach yielded CPR quality comparable to the conventional method. However, the authors cautioned that the foot-heel technique should be used carefully, as its effects on haemodynamics and the risk of resuscitation-related injuries remain unclear.
During FCCs, the rescuer stands and uses their leg, which may result in different compression forces compared to the kneeling position used in HCCs. As CPR involves rhythmic compression and decompression of the chest, differences in posture may affect the vertical alignment over the victim’s chest. This misalignment could lead to ineffective compressions or loss of balance, potentially causing injury to either the rescuer or the victim. Although previous studies2, 4, 5, 6 have considered these risks, no research has yet examined the force vector during FCCs.
Previous research has focused on the (bio)mechanical analysis of one- or two-handed CCs,9, 10 the mechanics of two-thumb or two-finger techniques in infant CCs,^11^ and the biomechanical characteristics of the chest during CPR,^12^ including gender-based differences in thoracic mechanical behaviour.^13^ To the best of the authors’ knowledge, no studies have investigated the (bio)mechanical characteristics of FCCs. The aim of this study was to compare the mechanical characteristics of leg-foot and two-hand chest compressions, to identify differences in quality and potential safety concerns between HCCs and FCCs.
This was a randomised crossover manikin study conducted between December 2024 and February 2025 at the University of Ljubljana, Faculty of Health Sciences. The reporting of this study adheres to the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guidelines for observational studies.^14^
In accordance with the internal standards of the Faculty of Health Sciences of the University of Ljubljana, the ethical appropriateness of the study was first assessed by the Chair of Public Health, who conducted a preliminary review to ensure compliance with ethical principles. As the study was an evaluation without the collection of personal or sensitive data, the Chair decided that approval from the National Medical Ethics Committee of the Republic of Slovenia was not required and written authorisation from the Dean of the Faculty was sufficient. In accordance with this procedure, the study was authorised by the Dean (Ref. No. 570/2024_1). This procedure reflects institutional practise for low-risk research that does not compromise the rights and integrity of participants. All participants provided written informed consent. Individuals who wished to withdraw from the study during training were free to do so at any time. Participants were assured of anonymity and the confidentiality of their data. The study was conducted in accordance with the principles of the Declaration of Helsinki.
Data were collected using the following •Force Plate: A Kistler 9286 AA force plate with BioWare data acquisition software (version 4.0.0.0, Kistler Biomechanics, USA) was used. Data were analysed in JupyterLab using Python packages including NumPy, Pandas, and SciPy, with signal processing and multidimensional image processing modules.•Manikin: A Resusci Anne QCPR Skill Guide manikin (Laerdal, Norway) equipped with a standard spring requiring approximately 412 N (42 kg) of force to achieve a 50 mm compression depth.15, 16 The manikin was connected to a Samsung Galaxy A35 smartphone (South Korea), which recorded data via the QCPR Learner app (version 7.0.0, Laerdal Medical, Norway).
The manikin was placed on the force plate in a standardised position (see Fig. 1) to ensure consistent and comparable 3D force data during CPR. The sampling rate of the force plate was set at 200 Hz.Fig. 1Snapshot of foot chest compressions (FCCs) being performed on the Resusci Anne QCPR manikin positioned on the Kistler 9286 AA force plate. The setup illustrates the standardised placement used to capture three-dimensional force data during compressions.
To relate the force plate data to chest displacement obtained from the QCPR Learner app, the compression phase of the mean force–displacement curve—describing the mechanical characteristics of the human chest during CPR^12^—was approximated using a linear (1)F[N]=(412N/50mm)×d[mm],where F is force in Newtons measured by the force plate, and d is the chest displacement in millimetres during the compression phase of the CPR cycle. The slope of the function represents the spring constant of the manikin, which is 412 N for 50 mm of displacement.^16^
Within the optimal CPR chest displacement range of dlow = 50 mm to dhigh = 60 mm,^7^ Eq. (1) yields corresponding force values of Flow = 412 N and Fhigh = 494 N, respectively.
It is important to note that the measured force represents the reaction force exerted by the manikin on the force plate, rather than the direct force applied by the rescuer’s hands or foot (see Fig. 2).Fig. 2Schematic representation of foot chest compressions (1) applied to a manikin (2) positioned on a force plate 3. The measured force vector F and its deviation from the vertical axis, represented by the polar angle θ, are illustrated in red. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
Eligible participants were healthcare students enrolled in practical first aid training at the University of Ljubljana, Faculty of Health Sciences, during the 2024–2025 academic year. Inclusion criteria providing written informed consent, being at least 18 years of age, and having the physical ability to perform basic adult life support on a manikin. Participants were randomly assigned to two Group (1) (n = 30), who performed hand chest compressions (HCCs) first, followed by foot chest compressions (FCCs); and Group 2 (n = 31), who performed the techniques in the reverse order. Group allocation was determined by drawing lots, with half of the numbers assigned to each group.
The study consisted of three parts. In Part 1, the quality of HCCs was measured; in Part 2, FCCs were assessed; and in Part 3, demographic data and participants’ subjective evaluations of their experience with both techniques were collected (see Fig. 3).Fig. 3Flow diagram illustrating the study design.
Before testing, participants practised HCCs on manikins. Each participant then performed both HCC and FCC in a randomised order. A rest period of 10 min was provided between the two phases, with the option to extend the break if needed. Each compression technique was performed for 1 min on the Resusci Anne QCPR Skill Guide manikin.
In the first part of the study, each student received clear instructions regarding the purpose of the research and was presented with a scenario in which they would perform chest compressions (CCs). The script “You are on the third floor of the University of Ljubljana, Faculty of Health Sciences. As you enter the hallway of office K313, you notice a man, approximately 40 years old, lying on the floor. Upon approaching him, you realise he is unresponsive and not breathing. There is no AED available. Your friend is already calling emergency medical services. Your task is to perform hand chest compressions (HCCs) WITHOUT rescue breaths.”
In the second part of the study, students received the same scenario, with the exception that they were instructed to perform CCs using their foot. They were asked to imagine they had an injured hand and could not perform HCCs. Participants were instructed to remove their slippers and perform CCs in socks to eliminate the influence of different footwear. No aids (e.g. chair, stool, or wall) were used during FCCs.
Immediately after completing the second part, demographic data and subjective assessments of the experience of performing both FCCs and HCCs were collected.
The primary outcomes included the time course of the force (magnitude and direction) exerted by the manikin on the force plate during resuscitation, as well as components of high-quality CCs,^7^ defined •Compression 50–60 mm.•Compression 100–120 compressions per minute.•Total number of compressions.•Chest 0 % to 100 %.•Duration of pauses in compressions.
These metrics were incorporated into the overall QCPR score, ranging from 0 % to 100 %. Further details on the QCPR scoring algorithm are available on the Laerdal Medical website (Laerdal Medical, Norway).^17^
Statistical analysis was conducted using IBM® SPSS® Statistics (version 29), Microsoft Excel (2007), and the SciPy statistical module in Python via JupyterLab. Variables were reported as mean ± standard deviation or median with interquartile range, depending on distribution.
The Shapiro–Wilk test was used to assess normality. For normally distributed data, a paired samples t-test was applied; for non-normally distributed data, the Wilcoxon signed-rank test was used. These tests were employed to compare the number of points achieved in FCCs and HCCs, as well as subjective difficulty ratings. The significance level was set at p < 0.05.
Sample size calculation was performed using G*Power.^18^ A total of 35 participants was required to achieve 80 % power with an effect size of 0.5 at an alpha level of 0.05. Accounting for an estimated 10 % attrition rate, a minimum of 40 participants was deemed necessary. However, the survey remained open beyond this threshold due to continued student interest in participation.
A total of 66 students participated in the study, of whom 61 were included in the final analysis. Two participants were excluded due to errors in the execution of the experiment (e.g. technical issues with time measurement) or incomplete data. Three students declined to participate, citing reasons such as illness (e.g. cold), injury (e.g. knee or wrist), or a medical condition (e.g. anorexia). Of all participants, the majority were female (n = 58), with ages ranging from 17 to 22 years. Approximately three-quarters were occupational therapy students (n = 33), followed by physiotherapy students (n = 22) and dental laboratory prosthetics students (n = 6). All participants had previously practised adult basic life support as part of their coursework at the University of Ljubljana, Faculty of Health Sciences.
To illustrate the data obtained from the force plate during chest compressions, Fig. 4 presents the amplitude of the measured force (F) and its polar angle (θ) over a time interval of several seconds, recorded from the same participant performing both HCCs and FCCs.Fig. 4ABOVE: Measured amplitude of the force (F) during several hand chest compressions (HCCs, red) and foot chest compressions (FCCs, black) performed by the same participant between approximately 6 and 8.5 s of the one-minute test. Transparent dots represent individual force measurements, while solid dots indicate the peak force for each compression. Dashed blue lines mark the thresholds for optimal CPR chest Flow = 412 N and Fhigh = 494 N, corresponding to 50 mm and 60 mm of chest displacement, respectively. BELOW: Corresponding polar angle θ of the force for values above the threshold Flow = 412 N. As above, transparent dots represent all measured values, and solid dots indicate the angle at peak force for each compression. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
During this interval, the maximal magnitude of the force applied for each FCC was lower than that for each HCC, although all maxima remained within the range corresponding to the optimal CPR chest displacement (50–60 mm). The difference in peak force between HCCs and FCCs ranged from approximately 10 % to 20 %. However, the difference in the polar angle θ exceeded 100 %. At the point of maximum force during the compression cycles, the deviation of the force vector from the vertical axis was greater than 10° for FCCs, compared to less than 5° for HCCs.
As illustrated in Fig. 4 for a specific time interval and participant, the statistical analysis of all measurements (n = 61 for HCCs and n = 61 for FCCs) revealed similar differences in maximal forces (Fig. 5) and corresponding polar angles (Fig. 6) between the two compression techniques.Fig. 5Histograms showing the frequency distributions of maximum force values (Fmax) during HCCs (BLUE) and FCCs (ORANGE) recorded during hand chest compressions (HCCs, blue) and foot chest compressions (FCCs, orange). Vertical dashed lines indicate the mean Fmax for each technique. The legend displays the corresponding mean (µ) and standard deviation (σ) values for both HCCs and FCCs. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)Fig. 6Histograms showing the frequency distribution of the polar angle (θ) at the point of maximum force (Fmax) during hand chest compressions (HCCs, blue) and foot chest compressions (FCCs, orange). Vertical dashed lines indicate the mean θ values for each technique. The legend displays the corresponding mean (µ) and standard deviation (σ) for both HCCs and FCCs. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
Fig. 5 presents histograms showing the distribution of the peak force during compressions (Fmax) for HCCs and for FCCs. The results of the statistical analysis, conducted using a paired t-test, indicate a significantly lower Fmax during FCCs compared to HCCs. The normality of the differences was confirmed (Shapiro–Wilk test, p = 0.42) and the paired t-test yielded a p-value of 0.01, a test statistic of 2.66, and a Cohen’s d of 0.34, suggesting a small to moderate effect size.
While the maximal force (Fmax) during FCCs is approximately 5 % lower than during HCCs, the deviation of the force vector from the vertical axis—expressed as the polar angle (θ)—is nearly twice as large for FCCs. Fig. 6 presents histograms showing the distribution of θ at the point of Fmax during the one-minute tests for both HCCs and FCCs. Statistical analysis using a paired t-test (normality of differences confirmed, p = 0.38) revealed a significantly greater polar angle for FCCs compared to HCCs (p < 0.0001, test statistic = –4.96, Cohen’s d = –0.63), indicating a moderate to large effect size.
Participants demonstrated significantly lower average CC depth while performing FCCs compared to HCCs (62 ± 12 mm vs. HCC: 69 ± 10 mm; p < 0.001) (Table 1), consistent with the lower maximal force observed (see Fig. 4). The compression rate was also significantly slower during FCCs (110 ± 15 CCs/min) than during HCCs (116 ± 18 CCs/min; p < 0.001). Similarly, the total number of compressions performed differed significantly between the two methods (FCCs: 110 ± 17 vs. HCCs: 116 ± 15; p < 0.001). The median chest recoil during FCCs was 96.0 % (IQR: 71.0–100.0), compared to 100.0 % (IQR: 81.5–100.0) during HCCs; however, this difference was not statistically significant (p = 0.126). Consequently, the median QCPR score was significantly lower for FCCs (87.0 %, IQR: 70.5–97.0) than for HCCs (97.0 %, IQR: 79.5–99.0; p = 0.043). Since no pauses occurred during the one-minute compression scenarios, the chest compression fraction was 100 % for both methods.Table 1A comparison of the two methods measured with Leardal QCPR.Components of the high-quality CCsHAND CCFOOT CC****Test statisticsDepth: M ± SD (mm)69 ± 1062 ± 12p* < 0.001Depth: MED (IQR) (%)100 (100–100)98 (85–100)p < 0.001Rate: M ± SD (CC/min)116 ± 18110 ± 15p < 0.001Rate: MED (IQR) (%)50 (5.5–91)41 (5.5–91)p = 0.987Recoil: MED (IQR) (%)100 (82–100)96 (71–100)p = 0.126Number of CC: M ± SD117 ± 15110 ± 17p < 0.001Abbreviation: M = Mean, MED = Median; SD = Standard deviation; IQR = Interquartile range; CCs = Chest compressions; *Wilcoxon signed-rank test.
This study investigated the mechanical differences between foot chest compressions (FCCs) and hand chest compressions (HCCs). The results demonstrate that the method of compression influences both the magnitude of force applied and its directional alignment. On average, HCCs produced greater force and deeper compressions than FCCs. For example, with an HCC mean peak force of 620 N, participants achieved a compression depth of 69 mm—exceeding current CPR guideline recommendations.^7^ In contrast, peak force during FCCs was approximately 5 % lower, and the force vector deviated more substantially from the vertical axis.
When evaluated against the recommended criteria for high-quality chest compressions,^7^ HCCs were slightly more effective in achieving adequate compression depth, rate, and vertical force. Consequently, FCCs yielded significantly lower QCPR scores, primarily due to reduced average compression depth. Minor deviations in recoil and rate may have also contributed. Nevertheless, both methods produced mean values close to guideline thresholds, suggesting that under simulated conditions, FCCs can still deliver acceptable compression performance.
Despite the lower quality metrics observed in the FCC group, adequate depth, rate, and recoil were still achievable. However, the higher standard deviation in FCC performance indicates more frequent deviations below the minimum recommended depth. This aligns with findings from related studies, which reported that participants often failed to meet the 2021 ERC adult BLS guideline target of 5–6 cm compression depth.3, 5, 6, 8, 19, 20 Regarding other quality indicators—such as compression frequency,8, 19, 4, 5, 6 recoil,4, 5, 19 and correct hand/foot placement4, 5, 19—our findings are consistent with previous research. Notably, our study recorded no interruptions in compressions, resulting in a 100 % compression fraction for both methods. Degel et al.^8^ reported slightly lower 99.8 % for HCCs and 98.8 % for FCCs.
The effectiveness of FCCs must also be considered in the context of specific circumstances. Physical limitations and rescuer fatigue are common barriers to high-quality HCCs, making the leg-heel approach a viable alternative.^8^ Trenkamp et al.^2^ found that individuals with arm or hand impairments tolerated FCCs better. Other studies3, 6 reported that FCCs may be more effective for individuals weighing less than 50 kg. Ott et al.^5^ highlighted the potential utility of FCCs during the COVID-19 pandemic, where increased physical distance from the patient was desirable.
Real-world studies have shown that even guideline-compliant HCCs can result in rib fractures (up to 95 %) and sternum fractures (over 60 %).^21^ Lung injuries are also common and may contribute to acute lung injury, complicating recovery.^22^ Magliocca et al.^23^ found that more forceful chest compressions (as applied during mechanical CPR compared to standard HCCs) are associated with a higher incidence of rib and sternal fractures, as well as CPR-associated lung edema (CRALE), which is characterized by reduced oxygenation and increased pulmonary shunt. Nevertheless, the higher cardiac output achieved with these stronger compressions ultimately results in greater oxygen delivery, highlighting the trade-off between increased injury risk and improved circulatory effectiveness.^24^ Considering these established risks associated with standard HCCs, safety concerns regarding FCCs must also be considered. Our findings challenge previous research2, 4, 5, 6 that raised concerns about increased injury risk due to higher forces generated by the leg. While manikin-based studies cannot fully assess injury potential, it is noteworthy that HCCs produced higher maximum forces, which may pose a greater risk of skeletal or visceral injury. FCCs may pose unique risks due to greater deviation from vertical force application. Tsou et al.,^10^ in a study comparing one-handed and two-handed compressions in children, noted that force directed toward the head could increase the risk of fractures or other injuries. Our study found a statistically significant difference in vertical deviation between FCCs and HCCs, but whether a 5 % difference translates into clinically meaningful outcomes remains uncertain. These risks must be evaluated in clinical or cadaveric studies, as manikin models are insufficient for injury assessment.
Several authors5, 20, 25, 26, 27 have proposed modifications to improve the safety and effectiveness of FCCs. Stabilisation aids such as chairs, footstools, or walls may help reduce imbalance and injury risk. Degel et al.^8^ also recommended the development of training protocols for FCCs as a second-line technique. We support the inclusion of FCC training in CPR courses, particularly for specific populations or in scenarios where alternative techniques may be necessary. Overcoming psychological barriers—such as reluctance to step on a victim—will be essential in promoting FCCs as a life-saving option.
This study was conducted using a manikin model, which limits the ability to assess actual clinical outcomes or evaluate the risk of resuscitation-related injuries. Such risks must be investigated in clinical settings or cadaver studies. Notably, the manikin’s rigid, flat back surface likely provides greater stabilisation during chest compressions than a real human body would, particularly on soft or uneven surfaces. This may influence mechanical feedback and force distribution, thereby limiting the generalisability of our findings to real-life scenarios.
The study sample consisted primarily of healthcare students, most of whom were young women. As such, the results may not be representative of the general population. Future research should explore the feasibility and effectiveness of FCCs among laypersons, particularly those with physical disabilities.
Participants did not receive formal training in FCCs prior to the study, which may have influenced performance outcomes. It remains unclear whether additional instruction and practice would have improved the quality of FCCs. Furthermore, the force plate system used in this study measured the pressure exerted by the manikin on the plate, rather than the direct force applied by the rescuer, which may introduce some limitations in interpreting the applied force dynamics.
Another limitation is the short duration of each chest compression trial, which was limited to one minute. While this allowed for standardised assessment of initial performance without the confounding effects of fatigue, it does not reflect the longer durations typically required during real-life CPR. Future studies should investigate performance over extended time intervals to better simulate clinical conditions.
This simulation study, focused on the mechanical analysis of chest compressions, revealed significantly lower compression quality metrics with foot chest compressions. Nevertheless, participants in both groups achieved compression parameters close to the guideline-recommended ranges, indicating that foot chest compression could be an acceptable alternative to standard hand compressions in specific circumstances.
Damjan Slabe: Resources, Project administration, Funding acquisition, Conceptualization. Eva Dolenc Šparovec: Writing – review & editing, Writing – original draft, Validation, Software, Methodology, Investigation, Formal analysis, Data curation, Conceptualization. Miha Fošnarič: Writing – review & editing, Writing – original draft, Visualization, Validation, Software, Methodology, Formal analysis, Data curation, Conceptualization.
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.