Abstract
Background
Progressive collapsing foot deformity (PCFD) represents a progressive medial longitudinal arch collapse of the foot. This study reviewed the clinical and biomechanical efficacy of orthotic interventions for PCFD.
Methods
A systematic search was performed on Web of Science, PubMed, and Scopus, resulting in 14 studies. We used the ROBINS-I tool to evaluate the methodological quality. For the narrative synthesis, we summarized the results in a textual format, while for the meta-analysis, Standardized Mean Differences were computed from the results whenever feasible.
Results
Major improvements were recorded in disability scores, pain intensity, and patient satisfaction. Orthotics decreased hindfoot eversion by 3–5° and increased forefoot plantarflexion by 6–9°. An ankle-foot orthosis with lateral extensions or an articulated design enhanced forefoot adduction and hindfoot inversion. Meta-analysis results indicate a 56.84 % improvement in the Foot Functional Index (95 % CI: 41.1–72.58) alongside a 28.53-point increase in the American Orthopedic Foot and Ankle Society Ankle/Hindfoot Scale (95 % CI: 22.66–34.4).
Conclusions
Orthotic interventions are effective treatments for PCFD, reducing pain, improving function, and correcting pathological kinematics.
Level of evidence
2
1
Introduction
The posterior tibial tendon (PTT) is a musculotendinous unit contributing to the ankle plantarflexion and foot inversion . PTT is a dynamic stabilizer of the medial longitudinal arch of the foot and hindfoot complex and provides crucial functions, including arch integrity and controlling pronation in the early stance phase of walking . Moreover, by stabilizing the transverse tarsal joints, the PTT helps with effective force transmission and propulsion in gait . Direct injury to the PTT along with a combination of mechanical overload, vascular insufficiency, acute trauma, subclinical collagenopathies, or age-related tendinopathy may lead to progressive collapsing foot deformity (PCFD) or as previously named posterior tibial tendon dysfunction (PTTD) . Epidemiological studies have indicated that the prevalence of PCFD within the general population ranges from 3.3 % to 10 %. In reality, the prevalence is probably far underestimated because of its late diagnosis in most cases that occurs with already late-stage changes in structure and apparent symptoms .
In this tendinopathy, there is a diminished capacity to actively invert the foot. In parallel, the peroneal tendons are compensatory activated , creating an abduction-eversion moment at the forefoot-midfoot. This kinetic creates tensile stress on the plantar fascia, and plantar medial ligaments of the foot, dominantly on spring ligament, thereby contributing to the progressive collapse of the medial longitudinal arch of the foot , . Consequently, the talus bone undergoes medial displacement and plantarflexion relative to the calcaneus, which in turn causes a valgus angulation of the calcaneus and a lateral deviation of the Achilles tendon with shortening , . PCFD patients notably demonstrate hindfoot valgus and plantarflexion, midfoot collapse, and forefoot abduction, pronation, and dorsiflexion deformities .
Conservative management strategies comprise the use of cold compresses, non-steroidal anti-inflammatory drugs (NSAIDs), custom orthotics, shoe modifications, therapeutic exercises, and ultrasound therapy . Orthotics are used to manage foot deformities and provide mechanical support to the posterior tibialis muscle ,, . This systematic review attempts to synthesize current literature regarding the efficacy of orthotic devices in the treatment of PCFD with an emphasis on patient-reported outcomes as well as objective biomechanical measures.
2
Method
This systematic review was conducted in alignment with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines for selecting studies and the PRISMA statement in rehabilitation guidance for accurate reporting . This research was prospectively registered in the International Prospective Register of Systematic Reviews PROSPERO (registration code: CRD42025637357).
2.1
Information sources and search strategy
The electronic search was finalized in January 2025 across the following bibliographic databases: Web of Science, PubMed, and Scopus. The search period was restricted to the years 2000–2025. There were no restrictions based on language. Table 1 provides a comprehensive overview of the search strategies employed across all three databases. The keywords were selected from Medical Subject Headings terms. Also, the reference lists of the included studies were examined manually to identify other eligible studies.
Table 1
Search strategy.
| Data Base | Query | Result |
|---|---|---|
| PubMed | (PCFD [Title/Abstract] OR “progressive collapsing foot deformity”[Title/Abstract] OR “Adult Acquired Flatfoot”[Title/Abstract] OR “Adult’s Pes Planus”[Title/Abstract] OR “Adult’s Pes Valgus”[Title/Abstract] OR “Posterior Tibial Tendon Dysfunction”[Title/Abstract] OR “Posterior Tibial Tendonitis”[Title/Abstract] OR “Posterior tibial tendon insufficiency”[Title/Abstract] OR “Adult Onset Flatfoot”[Title/Abstract]) AND (“Non-surgical”[Title/Abstract] OR “Orthos”[Title/Abstract] OR “Orthotic”[Title/Abstract] OR “Nonoperative”[Title/Abstract] OR “Brac”[Title/Abstract] OR “Conservative”[Title/Abstract]) | 90 |
| Scopus | (TITLE-ABS (PCFD OR “progressive collapsing foot deformity ” OR “Adult Acquired Flatfoot” OR “Adult’s Pes Planus” OR “Adult’s Pes Valgus” OR “Posterior Tibial Tendon Dysfunction” OR “Posterior Tibial Tendonitis” OR “Posterior tibial tendon insufficiency” OR “Adult-Onset Flatfoot”)) AND (TITLE-ABS (“Non-surgical” OR “Orthos” OR “Orthotic” OR “Nonoperative” OR “Brac*” OR “Conservative”)) | 120 |
| Web of Science | TS= ((PCFD OR “progressive collapsing foot deformity ” OR “Adult Acquired Flatfoot” OR “Adult’s Pes Planus” OR “Adult’s Pes Valgus” OR “Posterior Tibial Tendon Dysfunction” OR “Posterior Tibial Tendonitis” OR “Posterior tibial tendon insufficiency” OR “Adult-Onset Flatfoot”) AND (“Non-surgical” OR “Orthos” OR “Orthotic” OR “Nonoperative” OR “Brac*” OR “Conservative”)) | 105 |
2.2
Eligibility criteria
Eligibility criteria were defined for inclusion of studies based on PICOS items:
Inclusion criteria:
-
•
Participants: Studies included in this review focused on adult participants (aged 18–88 years) diagnosed with PCFD or adult acquired flatfoot due to PTTD in stage I, II, III, and IV based on the Johnson and Strom classification. A clinical diagnosis of flatfoot was acceptable for study inclusion.
-
•
Intervention: Studies included in this review examined the effects of foot, ankle, or foot and ankle complex orthotic devices.
-
•
Comparison: Trials evaluated one or more orthotic devices against either a control group, a placebo treatment, pre-intervention condition, or other rehabilitation interventions.
-
•
Outcome: Any outcome that can be attributed to the clinical and biomechanical outcomes of patients based on evidence.
-
•
Study design: This study incorporated Level II (randomized controlled trials) and Level III (pseudorandomized controlled trials with alternate allocation, comparative studies with concurrent controls, and comparative studies without concurrent controls) study designs as classified by the National Health and Medical Research Council’s level of evidence were included in this study .
Exclusion criteria:
-
•
Population: Research focusing on adult acquired flatfoot due to factors other than posterior tibial tendon dysfunction—such as tendon or ligament ruptures, along with conditions like Charcot arthropathy, rheumatoid arthritis, or neurological disorders, or cases without a defined cause—was excluded.
-
•
Intervention: Surgical, NSAIDs, and corticosteroids for initial symptom management were excluded.
-
•
Study design: Case reports, case series, cohort studies, editor correspondence, opinions, and technical notes were excluded.
2.3
Selection process
One reviewer performed a systematic literature search in the three mentioned electronic databases using the search queries described in Table 1 . After exporting the search results to EndNote software (20, Clarivate analytics, UK and USA) and removing duplicate studies, two reviewers independently evaluated the titles and abstracts of each identified record. Then, full texts of relevant studies were retrieved and assessed against the previously defined eligibility criteria. When clarification was required regarding study methodologies or data, or eligibility for inclusion was uncertain, the two reviewers discussed together until consensus was reached.
2.4
Data collection process
Data from included reports were extracted and abstracted by two reviewers. Each reviewer worked independently of the other. A standard data extraction form was used in Excel software (2019, Microsoft, USA). In instances when data were insufficiently clear, the corresponding reviewer contacted study investigators by email to request necessary information or to verify specific data points.
2.5
Data items
Two reviewers independently gathered data regarding the study characteristics and findings: authors’ names, year, study design, number and demographic characteristics of participants, interventions, duration of study, assessment tools, outcomes, and results. Mean difference ± SD were extracted and used in the synthesis or presentation of results for effect measures.
2.6
Study risk of bias assessment
Two reviewers evaluated each trial included in the study independently, utilizing the online ROBINS-1 (visualization tool) Risk of Bias tool ( https://www.riskofbias.info/welcome/robvis-visualization-tool ) across the subsequent seven domains: bias due to confounding, bias due to selection of participants, bias in classification of interventions, bias due to deviations from intended interventions, bias due to missing data, bias in measurement of outcomes, and bias in selection of reported results . For every one of these seven domains, we used a classification of Critical, Serious, Moderate, Low, and No information according to the standards set forth in the Cochrane Handbook . Cohen’s kappa was utilized to assess the inter-rater reliability between the evaluations conducted by the two reviewers, with any conflicts addressed through discussion.
2.7
Synthesis methods
Narrative and quantitative methods were used for the synthesis. For the narrative synthesis, we summarized the characteristics of the studies, the interventions, and the reported outcomes of included studies in textual format. Meta-analysis was conducted using Review Manager (Cochrane Collaboration, UK, v.3.5). Statistical results were reported as standardized mean differences and 95 % CI. Results were regarded as homogeneous if P > 0.05 and I² < 50 %, and as heterogenous if P ≤ 0.05 and I² ≥ 50 %. Homogeneous data were examined using a fixed effects model, whereas heterogeneous data were assessed with a random effects model. We also sought to evaluate publication bias through funnel plots and the Egger regression test. P < 0.05 defined statistically significant differences .
3
Results
We identified 284 studies from an electronic database search. After removing 78 duplicates, 206 studies were screened, of which 167 were considered ineligible. Eligible studies were identified based on full text review of 39 quantitative studies, resulting in the exclusion of 25 additional studies. The final review included 14 studies ,,,,,,,,,,,,, ( Fig. 1 ).
PRISMA flow diagram of study selection.
3.1
Methodology quality assessment
The risk of bias assessment of included studies is presented in Figs. 2 and 3 . The two reviewers scored 98 criteria independently, with 89 of the responses matching, indicating an inter-rater agreement rate of 90 %. The Cohen’s kappa index of inter-rater reliability yielded a k-coefficient of 0.85 (95 % CI: 0.73, 0.96). The methodological quality evaluation of the included studies revealed that confounding was at moderate risk in 10 studies ,,,,,,,,, , low risk in 2 studies , , and serious risk in 2 studies , . Moreover, the risk of bias assessment of the participant selection domain was low for 8 studies ,,,,,,, , unclear for 1 study , critical for 1 study , moderate for 2 studies , , and serious for 2 studies , . Bias assessment for classification of interventions was also found out as low in 9 studies ,,,,,,,, , as moderate in 3 studies ,, , and serious in 2 studies , . The risk of bias assessment for deviations from the intended interventions indicated that 10 studies had low risk ,,,,,,,,, , 3 had moderate risk ,, , and 1 study had a serious risk . More so, bias due to missing data was assessed as low in 8 studies ,,,,,, , moderate in 5 studies ,,,, , and serious in 1 study . The assessment of bias in the measurement of outcomes revealed that 5 studies had a low risk ,,,,, , 6 studies had a moderate risk ,,,,, , and 2 studies had a serious risk , . Lastly, bias in the selection of reported results was evaluated as low in 6 studies ,,,,, , serious in 1 study , and moderate in 7 studies ,,,,,, .
Traffic light plot of quality assessment result with ROBINS-I tools.
Summary plot bias weighing result.
3.2
Study characteristics
Table 2 displays the characteristics of the studies. Study designs included RCTs ,,,, , prospective observational studies , pre-post intervention analyses ,,,, , controlled laboratory experiments ,,, , and case series . Demographic participant characteristics varied among studies: sample sizes ranged from 10 to 67 participants and mean ages from 30.66 to 64.2 years. Interventions were from orthotic management, including prefabricated ,, and custom FOs ,,, , Ankle Foot Orthoses (AFOs) ,,, , and specific designs like the Dynamic University of Arizona Foot Orthosis (DUAFO) , and Computer aid design-computer aid manufacturing (CAD-CAM) orthoses . Assessment tools were: American Orthopedic Foot and Ankle Society (AOFAS) Ankle/Hindfoot Scale ,, , Foot Function Index (FFI) percentage ,,, , Visual Analogue Scale for pain ,, , and Short Musculoskeletal Function Assessment . Further objective measurement instruments used include the motion-analysis systems ,,, . These study durations have ranged from immediate laboratory assessments ,,, to long-term follow-ups as long as 7 years , enabling both the immediate biomechanical effects and the sustained clinical outcomes of such treatments to be evaluated. The key outcomes reported include pain reduction ,,,,,,, , functional improvement ,,,,,,,, , alteration in kinematic parameters , and patient satisfaction ,, .
Table 2
Details of study characteristics.
| Author, Year | Design | Participants | Intervention | Outcomes and Assessment Tools | Duration | Results |
|---|---|---|---|---|---|---|
| Clinical outcomes | ||||||
| Jari et al., 2002 | pre- and post-intervention |
N:38
Male/female:21/7 Mean age: 60years |
Stage 1 and 2 anti-inflammatory medication, ice, mega pulse, and ultrasound, strengthening exercise, UCBL orthoses
Stage 3 and 4: AFO |
-The AOFAS for function assessment | 7–12 month | All patients assessed expressed satisfaction with their treatment results, and only one patient in stage 1 has asked for surgery. |
| Bek et al., 2003 | Single-group pre- and post-intervention |
N:25
Male/female:10/15 Mean age: 30.66 years |
Medial wedge (N = 4)
Medial wedge and medial arch support (N = 8) Medial arch support (N = 4) UCBL (N = 6) Air cast (N = 2) Rigid mold (N = 1) |
-The VAS for pain assessment
-FFI |
6 weeks | Orthotic intervention will not alter the inherent deformity or the lack of muscle strength, but by providing support to the foot, it may help alleviate pain and potentially lower the risk of disability. |
| Alvarez et al., 2006 | A prospective, observational study |
N:47
Male/female:10/37 Mean age: 50.3 years |
A short articulated AFO
Three-quarter length FO Exercise program |
-To assess pain, VAS was used.-Function was evaluated by measuring the ability to walk a specific distance.-Strength was assessed by measuring the isokinetic strength of the ankle muscles using a dynamometer. | 3 years | On average, after about 4 months, most patients had made enough improvement to be able to function normally. |
| Lin et al., 2008 | Pre-post intervention |
N:32
Male/female:5/27 Mean age: 57.6 years |
The DUAFO consisted of
a steel shank, a full rocker bottoms sole, and a medial T-strap |
-AOFAS Ankle/Hindfoot Score
-SF−36 -FFI -VAS for pain |
7 years | The mean AOFAS and FFI scores were improved. The SF−36 subscores across all age categories showed no notable differences (p > 0.05). Additionally, there was an increase in the satisfaction rate. |
| Houck et al., 2015 | Prospective randomized study |
N:39
Male/female:8/31 Mean age: 57.5 years |
Air cast orthosis consists of ankle stirrup and medial arch support. | -FFI
-Short Musculoskeletal Function Assessment -the strength of the isometric deep posterior compartment with dynamometer |
12 weeks | The use of orthoses alone in participants showed limited effectiveness for stage II PTTD. |
| Kulig et al., 2009 | Prospective randomized study |
N:36
Male/female:15/21 Mean age: 52.3 years |
FOs (n = 12)
FOs and concentric exercise (n = 12) FOs and eccentric exercise (n = 12) |
-FFI
-Distance traveled in the 5-Minute Walk Test |
12 weeks | Post-intervention, FFI scores (overall, pain, and disability) demonstrated a reduction across all groups. The cohort that underwent both orthosis and exercise displayed the most significant improvement in every category, whereas the Orthosis-only group experienced the smallest advancement. Furthermore, pain levels immediately after the 5-Minute Walk Test notably diminished. |
| Yurt et al., 2018 | RCTs |
N:67
Male/female:45/21 Mean age: 58/2 years |
Computer aid design-computer aid manufacturing FOs (n = 22)
Conventional FOs (n = 22) Sham FOs (n = 22) |
Pain with VAS
Health related quality of life |
2 months | Both computer-aided design and computer-aided manufacturing of FOs, as well as conventionally designed FOs, are more effective at alleviating pain than sham FOs. |
| Esterman et al., 2005 | Pilot, randomized controlled trial |
N:47
Male/female:44/3 Mean age:60.1 years |
Three quarter length Australian orthotics Laboratory flexible FOs | -The Foot Health Status Questionnaire
-Health Organization Quality of Life Questionnaire |
8 weeks | Although the results were not statistically significant, participants who utilized orthotics reported the lowest levels of lower limb pain, as well as the best overall foot health and quality of life. |
| Koltak et al., 2021 | Pre-post intervention |
N:52
Male/female: 27/25 Range age: 18–60 years |
FOs were created using a computer-assisted modeling process that incorporated medial longitudinal arch supports (ranging from 8 to 12 mm) and transverse arch supports (ranging from 4 to 6 mm) for all participants. Additionally, 6-degree medial heel wedge pads were applied for those with calcaneal eversion. |
FFI
Muscle strength: using an isokinetic dynamometer (Humac Norm, CSMi, USA). |
9 months | A significant enhancement was detected in the FFI (p < 0.05) |
| Krause et al., 2008 | Prospective case series (pre- and post-intervention) |
N = 18
Male/Female:4/14 mean age: 64.2 years |
The custom-molded FO was designed by positioning the patient with their knee flexed at a 90-degree angle. The various aspects of the deformity were manually adjusted, and a cast was created from the foot in this position. The sole of the FO was crafted from 2-mm polypropylene, while the shell was made from a more flexible rubber that extended 1–2 cm above the malleolar level for added support. | The AOFAS ankle hindfoot score and clinical or radiographic progression | 61.4 months | Among the 18 patients, three (16 %) showed clinical progression to a stable deformity (Stage III) along with a deterioration in their radiographic findings. The other patients expressed contentment with the comfort of the FO and observed enhancements in their mobility. Additionally, complications occurred in three patients (16 %), involving callus formation. |
| Biomechanical outcomes | ||||||
| Neville et al., 2016 | Controlled laboratory, repeated measured |
N:15
Male/female:9/6 Mean age: 51.8 years |
A standard AFO, an AFO with a lateral extension, and a shoe-only control condition |
A motion-analysis setup 12 Vicon 512 cameras and Workstation Version 5.2 software was used for data capturing.
The outcomes were: -hind foot inversion/eversion -forefoot plantar flexion/dorsiflexion- forefoot abduction/adduction |
Immediate | An AFO with a lateral extension significantly improved forefoot adduction in individuals with PCFD while walking. |
| Neville et al., 2009 | Experimental laboratory study |
N = 10
Male/ Female:6/4 Mean age: 54.8 |
A dose response orthosis with air bladder inflation to 0, 4, and 7 PSI. |
Two arrays of infrared cameras were employed in conjunction with Motion Monitor software.
Kinematic data were gathered from the tibia, calcaneus (hindfoot), and first metatarsal. |
Immediate | The most notable rise in forefoot plantar flexion took place in the third rocker. The greatest variation in adduction was observed during the third rocker. |
| Neville et al., 2012 | Experimental laboratory study, repeated measured design |
N = 15
Male/ Female:7/8 Mean age: 53.8 |
Shoe only (control condition), shoe with a custom solid AFO, shoe with a custom articulated
AFO and shoe with a prefabricated AFO |
A 12-camera VICON motion capture system, paired with Workstation software.
Kinematic data were: -hindfoot inversion, -forefoot plantarflexion -forefoot adduction |
Immediate | The custom AFO exhibited considerably greater hindfoot inversion during the loading response (p = 0.002), mid-stance (p < 0.001), and terminal stance (p = 0.02). Furthermore, the custom solid device, and the prefabricated available device caused increased forefoot plantarflexion relative to the shoe-only. |
| Chicoine et al., 2020 | Experimental laboratory study, repeated measured design |
N = 14
Male/ Female:8/6 Mean age: 55.6 |
Shoes, prefabricated FO, neutral custom FO, and custom FO with a 5° medial wedge and a 4 mm medial heel skive. |
A three-dimensional motion analysis system with 10 cameras.
The outcomes were: The motion of lower limb joints, and joint moments. |
Immediate | Custom foot orthotics decreased hindfoot eversion angles, diminished ankle inversion moments, and increased ankle eversion moments in comparison to both the Shoe and PFO conditions (p < 0.001). Furthermore, both custom FOs and controlled variable FOs resulted in a greater knee adduction moment relative to the Shoe condition (p < 0.001). |
Related posts:
The effect of preoperative osteoporosis on postoperative outcomes following total ankle arthroplasty
Resection Arthroplasty of the Distal Radioulnar Joint
Thumb Carpometacarpal Joint Implant and Resurfacing Arthroplasty
Surgical Treatment of Vasospastic and Vaso-occlusive Diseases of the Hand
Perioperative Management of the Patient With Coronary Stents
Anterior Cruciate Ligament Reconstruction via the Anteromedial Portal and Single-Tunnel, Double-Bundle Techniques
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
