Effectiveness of high-intensity laser therapy in patients with plantar fasciitis: A systematic review with meta-analysis of randomized clinical trials

This SR was conducted following the guidelines established by the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA 2020) [18]. The SR was registered in the International Prospective Registry of Systematic Reviews (PROSPERO, registration number CRD42023388376) [19].

The research question followed the PICO (population, intervention, comparison, and outcome) approach. The population comprised patients diagnosed with PF, treated with HILT, and compared with those who received other physical therapy interventions, with or without placebo HILT. The main outcome was pain reduction using recognized instruments such as the visual analog scale (VAS), numerical scale (NPRS), or other validated scales. In addition, relevant secondary outcomes were considered, such as functionality assessment using widely accepted scales such as the Foot and Ankle Outcome Score (FAOS) or Foot Function Index (FFI).

An electronic search for randomized clinical trials (RCTs) evaluating HILT in FP was performed. Recognized databases were utilized and included PubMed, Web of Science (WoS), Scopus, CINAHL, Science Direct, and the Physiotherapy Evidence-Based Database (PEDro) (last update: July 23, 2023).

Searching was performed using a set of keywords selected from the Medical Subject Headings (MeSH) dictionary. Keywords included “lasers”, “laser therapy”, “phototherapy”, “high-intensity laser therapy”, “class IV laser”, “musculoskeletal pain”, “fasciitis”, “fasciitis, plantar”, “foot Diseases” and “heel Spur”. These terms were combined using the boolean connectors “OR” and “AND” to obtain the search algorithm: (((((“Lasers”) OR (“Laser Therapy”)) OR (“Photo-therapy”)) OR (“High Intensity Laser Therapy”)) OR (“Class IV laser”)) AND (((((“Musculoskeletal Pain”) OR (“Fasciitis”)) OR (“Fasciitis, Plantar”)) OR (“Foot Diseases”)) OR (“Heel Spur”)). In addition, the “clinical Trial” and “randomized controlled trial” filters were applied to ensure the inclusion of RCTs in the search.

The analysis of the titles and abstracts downloaded from each of the databases was carried out by three independent researchers (FJR-DPQ-FPM). To expedite this process, the Rayyan web tool was used [20], which facilitated an accurate assessment based on predefined selection criteria.

This review considered as inclusion criteria: a) human RCTs with PF diagnosis; b) treatment with HILT as either the sole intervention or in combination with other therapies; c) comparison with other physical therapy treatments or HILT placebo; d) the main outcome was centered on pain intensity changes. Literature reviews and other SRs on HILT, other foot musculoskeletal or neurological conditions, and studies with incomplete or unavailable texts were excluded.

The Cochrane Collaboration risk of bias (RoB) tool was used to determine bias in the studies that were included [21]. Studies that had two or more high risks of bias were of low quality. The kappa statistic was used to assess the agreement in the assessment of bias between the researchers [22].

The Review Manager software (RevMan) 5.4 was used for statistical analysis [23]. Heterogeneity between studies was assessed using the I² statistic in the categories negligible, moderate, substantial, or considerable [24]. The Mantel-Haenszel fixed-effects method was used to calculate the pooled mean difference for the interesting outcomes, with a 95% confidence interval.

The assessment of evidence quality was conducted through the utilization of the GRADE approach (Grading of Recommendations, Assessment, Development, and Evaluation) [25]. To provide a comprehensive summary of evidence concerning HILT in relation to PF, researchers employed the GRADEpro GDT tool for the purpose of guideline development (www.gradepro.org).

The initial search in the electronic databases yielded a preliminary list of 2.315 articles (PubMed = 18; Scopus = 325; WoS = 75; CINAHL = 89; Science Direct = 1.695; Cochrane Central = 108; and PEDro = 5). After eliminating duplicate articles, 741 documents were obtained for analysis. Initially, 16 articles were chosen, but seven were excluded due to their relevance to other foot conditions (plantar ulcer and Achilles tendinopathy), research on LLLT in PF, and an incomplete study. This left nine articles for analysis [26,27,28,29,30,31,32,33,34]. Figure 1 depicts the flowchart with the search strategy.

Fig. 1.

PRISMA flow chart diagram

Figure 2 presents the RoB assessment conducted by three investigators (HDB, FJR, CVI) for the studies encompassed within the analysis [26,27,28,29,30,31,32,33]. The analysis revealed commendable inter-rater agreement in evaluating bias (kappa 0.82) [22]. Elevated RoB levels were predominantly discerned in the context of randomization sequence (22.2%), concealed allocation (22.2%), and assessor blinding (55.6%). Conversely, the criteria of selective reporting and incomplete data demonstrated the least susceptibility to RoB.

Fig. 2.

Risk of bias summary: review author's judgements about each risk of bias item for each included study

Table 1 provides a comprehensive overview of the key attributes of the selected RCTs. Pertinent information encompassing study groups, participant selection criteria, interventions administered, assessments, and main outcomes is elucidated. These RCTs were conducted across diverse geographical locations, such as Lithuania, Turkey, Malaysia, Poland and Pakistan, spanning the period from 2019 to 2023. Overall, a total of 447 participants diagnosed with PF were enrolled, demonstrating an average age of 50.7 years (SD ± 10.6). Among these participants were 274 women, 128 men, and one study that did not provide specifications regarding sex. A total of 237 patients underwent HILT, while 210 controls (CGs) received conventional physical therapy. In the experimental group (EGs), 60 patients received only HILT [26,32,33], while 177 participants received HILT in combination with insoles [27,28,29], stretching exercises [28,29,30,34], therapeutic ultrasound (US) [31] or cryotherapy [29]. In CGs, LLLT [28,29], insole [27,28,29], US [31] and ECSWT [33,34] were used. In addition, two studies applied the HILT placebo [30,31].

HILT treatments were administered to the plantar surface, predominantly utilizing the scan technique with a 30-mm-diameter spacer [26,27,28,29,30]. In two studies, the punctual technique was employed [31,33], whereas in one study, the application method was not specified [32]. Neodymium-doped yttrium-aluminum-garnet (Nd:YAG) lasers with a wavelength of 1.064 nm were employed, featuring maximum powers of 3.000W and 7W in a pulsed emission mode, with an average power of 7W, and energy delivery spanning from 500 to 4496J. Additional HILT parameters, encompassing pulse frequencies (Hz), energy density (J/cm²) and treatment time, are outlined in Tab 1. Notably, treatment sessions ranged from six to fifteen, conducted over intervals of 3 to 4 weeks.

All investigations assessed pain intensity during rest, the first steps, and/or movement. The primary instruments utilized for this purpose were the VAS or FAOS [27,28,29,30]. Additional measured variables encompassed pain pressure threshold [26,29], range of motion [26], and disability assessment using either FAOS [27,28,30] or the foot function index (FFI) [32,33]. In addition, some studies considered changes in PF thickness using ultrasonography [26,29,33], quality of life (QoL) measured by the SF-36 questionnaire [30], and plantar contacts using podography [30]. All studies conducted assessments before and after treatment, and three of them conducted follow-up sessions between four and twelve weeks after treatment.

Table 2 summarizes the interesting outcomes of the RCTs that were included. Pain reduction was observed for each study in all groups (p < 0.05), both in the HILT-treated group and in the CGs, during the assessments [26,26,27,28,29,30,31,33]. However, at the end of the treatment and follow-up, greater analgesia was observed for the HILT groups. Disability shows statistically significant changes in both groups before and after treatment (p < 0.05), but the results were contradictory when it came to determining whether HILT was more effective than other physical therapy interventions [27,28,30,33]. Improvements were found in PF thickness [26,29,33] and in QoL in the EGs (p<0.05) [30], highlighting greater effectiveness for HILT.

Seven studies were considered for meta-analysis in relation to pain intensity (Fig. 3) and functionality (Fig. 4). The Mantel-Haenszel fixed effects method was used to determine the pooled effect by mean difference (MD) [23,24]. The results show a statistically significant difference in favor of HILT in pain intensity at rest (VAS: MD = − 0.70cm; 95% CI = −1.10, −0.30; p < 0.01) (FAOS: MD = 5.93%; 95% CI = −1.55, −0.68; p < 0.01), as well as a reduction in pain in the first steps (VAS: MD = −1.27cm; 95% CI = −1.87, −0.67; p < 0.01). However, no statistical differences were found in pain intensity when walking, pain intensity when sitting, when comparing rest pain versus ECSWT, pain at rest at 3-month follow-up. Furthermore, when comparing rest pain between HILT and LLLT added to a physical therapy plan, a statistically significant change was observed for HILT (VAS MD = −2.76cm; 95% CI = −3.51, −2.0; p < 0.01). The I² coefficient revealed a negligible or moderate degree of heterogeneity, except for pain at the first steps. Researchers assessed the quality of the evidence as important for pain intensity in first steps (VAS) and least pain at rest (FAOS), but with low certainty due to high inconsistency. However, comparing HILT and LLLT evidence quality was rated as critical and of high certainty (Tab. 3).

Fig. 3.

Forest plots for pain intensity assessed with VAS (3A-3G) and FAOS (3H) at the end of treatment

Fig. 4.

Forest plots for FAOS and FFI: (4A) HILT versus control for daily life activities (FAOS subscale); (4B) HILT versus control for symptoms (FAOS subscale); (4C) HILT versus control in performance of sports and recreation activities (FAOS subscale); (4D) HILT versus control for quality of life (FAOS subscale); (4E) disability with FFI comparing HILT versus ECSWT at the end of treatment

No statistical differences were found between the groups for FAOS in activities of daily living, symptoms and development of sports activities. Nevertheless, a significant difference in favor of HILT was observed for QoL (MD = 14.42%; 95% CI = 9.43,19.40; p < 0.01). There was also no difference in disability for FFI when HILT was compared to ECSWT. The heterogeneity obtained for disability was important, except for activities of daily living. The quality of the evidence for the improvement in QoL was considered important, but with low certainty due to inconsistency (Tab. 3).

HILT reduces pain at rest (VAS) when combined with interventions such as exercises, insoles, or cryotherapy, with an average decrease in pain of −0.7 cm (95% CI: −1.1, −0.3), results that, despite being statistically significant on paper, are not clinically important for VAS, where an average decrease of −0.9 to −1.3cm (95% CI: 0.6 to 1.8) is expected. [28,29,30,33,34,35,36]. Moreover, when compared to LLLT, HILT seems more effective, with an average analgesia of −2.76cm for VAS (95% CI: −3.51–2.0), although it should be noted that the number of RCTs that compared both lasers was limited (28,29), which could lead to publication bias and highlights the need for further research. In addition, the analgesic effects of HILT are notable in the first steps, with a pooled effect of −1.27cm for VAS (95% CI: −1.87, −0.67).

Despite the statistically significant results and the considerable effect size, the comparisons between HILT and LLLT are the most reliable due to the lack of heterogeneity in the data (I² = 0%). This is in contrast to pain at rest, as well as in the first steps or after walking, where a moderate-to-high level of heterogeneity is observed [24]. The result in the first steps is consistent with the established minimally important clinical difference (MCID) for VAS of at least −1.3cm [35]. Nevertheless, concerning pain during rest, the decrease of −0.7cm falls short of this value. This suggests a more pronounced analgesic effect of HILT in relation to functional tasks like walking, which retains significance due to its involvement in routine activities of daily life [27,29,33]. The decline in pain aligns congruently with outcomes reported in other SRs investigating LLLT for PF. They have exhibited an average pain reduction of approximately −1.3cm (95% CI: −0.4,–2.3) or −0.95cm (95% CI: −1.2, −0.7) when evaluating LLLT either alone or in conjunction with US, ECSWT, or exercises [13,14]. Hence, there exists substantiation to contemplate the adoption of both lasers for analgesic purposes. These conclusions have prompted researchers to assign a good level of evidence concerning the influence of HILT on pain at first steps, while deeming it of utmost critical importance when compared with LLLT for pain during rest.

Furthermore, there is significant evidence pointing to the heightened analgesic efficacy of HILT compared to LLLT, insoles, or placebo, as evidenced by the results in FAOS. However, even though this finding is statistically significant, the cumulative effect size of 5.9% (95% CI: 2.39,9.47) does not meet the recommended MCID threshold of 9.5% for the pain subscale [37].

The superior analgesia of HILT over LLLT can be attributable to quicker energy delivery, particularly in continuous mode, and the ability to cover larger treatment areas through scanning applications, as outlined in the RCTs [28,29]. This phenomenon mirrors the Reprocity Busen Roscoe's principle, wherein enhanced power results in quicker attainment of physiological effects (38). In addition, continuous emission allows for a greater thermal effect, giving HILT advantages over LLLT and the other treatments.

Although the analgesic effects of the laser are clear and are based on disminuting the inflammatory process, β-endorphins release, cytochrome C-oxidase activation, and decreasing nociceptors discharge, these effects could be potentiated by the generation of heat [11,12,17]. The elevation in temperature reinforces the analgesic impact by promoting vasodilation, thereby aiding in the removal of inflammatory mediators. This process reduces the activity of nociceptors and sensitizes transient receptor potential cation channel subfamily V member 1 (TRPV-1) receptors, particularly when the heat remains consistent, and the skin reaches a temperature of 37°C [39]. Moreover, muscle relaxation induced by heat could potentially contribute to an additional analgesic effect by promoting the disruption of the muscle spasm-pain cycle [40]. Furthermore, the rise in temperature also impacts tissue viscoelastic properties, a factor that is advantageous when considering the integration of laser therapy with stretching exercises, a practice strongly advocated for managing PF to alleviate pain and enhance functionality [8,41]. This aligns with the SRs of LLLT and PF, where RCTs incorporating stretching exercises demonstrated a notable decrease of −1.98cm in the VAS score post-treatment, resulting in both statistical and clinical significance [14]. This underscores the imperative of complementing laser therapy, irrespective of its type, with stretching exercises.

During the 3-month follow-up period, there were no discernible differences in resting pain between the study groups, suggesting that HILT may be more effective in the short term than the long term. Corresponding observations are noted for LLLT, where the analgesic effect is more pronounced during the treatment period [13,14]. This prompts the consideration of incorporating HILT into an intervention plan while concurrently exploring supplementary approaches to enhance post-treatment outcomes. For instance, implementing stretching exercises targeting the PF and gastrocnemius muscles could be valuable in optimizing results beyond the treatment period [8,41].

The superiority of HILT over ECSWT remains uncertain, a consideration of significance given the established evidence supporting ECSWT in cases of PF [33,34]. Analogous outcomes have emerged from comparisons between LLLT and ECSWT, revealing no statistically significant distinctions [13,14]. Consequently, this situates HILT as an equivalent or supplementary clinical option for PF treatment, potentially on par with ECSWT. The choice between these modalities could hinge on factors such as resource availability, cost considerations, and patient or physical therapist preferences.

The RCTs utilized FAOS as the principal instrument for evaluating functionality. FAOS is endorsed for its established validity in appraising plantar pain, with robust reliability across all its subscales (ICC: 0.81–0.92) [42,43]. FAOS assessments reveal that HILT does not demonstrate superiority over other interventions such as LLLT, insoles, or the combination of placebo HILT and exercises in terms of daily activities, symptoms, and sports activities [27,28,30]. Nonetheless, it is worth noting that HILT does exert an influence on QoL, manifesting an average enhancement of 14.4% in comparison to these treatment modalities. This outcome is of significance, as it exceeds the threshold recognized as the MCID for this subscale, established at 5% [37,43].

These findings are significant due to the well-established relation between pain and QoL, particularly in the domains of physical and emotional functioning [44]. Consequently, a treatment with a substantial analgesic effect is highly likely to exert an influence on QoL. However, this also depends on the duration, intensity, scope, affectivity, and pain meaning, which implies that this improvement will not always be a “sine qua non” condition [44,45]. Furthermore, it is crucial to understand that pain does not always indicate poor QoL, although it is a significant factor [44]. Despite being a social construction, QoL has also gained importance in health research as an interesting outcome, which makes measuring it relevant, especially when complaints are chronic [44,46].

Likewise, the QoL results should be analyzed in greater detail due to the I² index between the studies [24]. This consideration prompted researchers to categorize the evidence related to HILT's impact on QoL as important, though not of the highest level.

Similar findings in disability for FFI were obtained in two RCTs comparing HILT with ECSWT, with no significant change in favor of either treatment. However, this conclusion is not definitive due to the limited number of studies [33,34].

This SR reveals a marked variability of HILT dosages used among RCTs, a situation analogous to that of LLLT SRs [13,14]. However, the authors have established a dosage recommendation for the application of HILT at a wavelength of 1.064nm, with the following predefined parameters: an output power of 12W, continuous mode, an energy density of at least 120J/cm² and a minimum total energy of 3.000J for a treatment area of 25 cm². The scanning application was the predominant one, presumably to cover the entire sole of the foot. Moreover, sessions should range between 8 and 10, carried out at intervals for a period of no less than 3 weeks. Furthermore, the best results seem to be seen by adding PF and calf stretching to HILT and the use of insoles [27,28,30,34].

A single study reports HILT plus cryotherapy [29]. However, the researchers propose cryotherapy for the exacerbation of symptoms after activities like walking or towards the end of the day, thereby avoiding any potential interference with the thermal effects induced by HILT [47].

To our knowledge, this is the first SR to evaluate the effectiveness of HILT in FP. The transparent approach based on the PRISMA guidelines and the registration of the protocol in PROSPERO to evaluate and present the evidence are highlighted. Limitations identified by researchers include: (i) despite an extensive search across six databases, the potential inclusion of articles in languages beyond those covered cannot be definitively ruled out due to the geographic origin of the RCTs in Turkey, Malaysia, Pakistan, and Lithuania; (ii) the considerable heterogeneity among RCTs impedes our ability to provide a conclusive analysis for pain at rest and functionality, thereby limiting the definitive interpretation of these aspects; (iii) RoB in certain RCTs, particularly concerning the blinding of assessors and concealed allocation, raises the possibility of an overestimation of the effects attributed to HILT or conventional physiotherapy treatments.

The authors have outlined two potential future research directions. Firstly, they propose conducting comparative clinical trials to assess the effectiveness of HILT and ECSWT in the treatment of PF. Secondly, they recommend further investigations into HILT for PF, including an examination of the outcomes with additional treatment sessions of HILT and a direct comparison with Low-Level Laser Therapy (LLLT). It should be noted that the mean number of treatment sessions across the included randomized controlled trials (RCTs) ranged from 8 to 10. It is plausible that increasing the number of treatment sessions could enhance the long-term efficacy of HILT, given the chronic nature of PF.