Background
Chronic plantar heel pain (CPHP) is one of the most common conditions affecting the foot and has been reported to account for 15% of all adult foot complaints requiring professional care [
1]. It is usually observed in the 40 to 60 year old age bracket, but has been reported in people from 7 to 85 years and appears to be more common in females [
2]. Symptoms typically include pain under the medial heel during weight bearing, especially in the morning and at the beginning of weight-bearing activities [
1,
3].
As with many conditions where the true pathology is unclear, CPHP has become a generalised term encompassing a broad spectrum of conditions affecting the heel, including subcalcaneal bursitis, neuritis, plantar fasciitis and subcalcaneal spur [
4,
5]. However, plantar fasciitis is considered to be the most common cause of pain and the terms are used interchangeably in the literature [
1]. Due to the apparent heterogeneity in the conditions grouped together as CPHP, it is difficult to determine a definitive aetiology for the condition [
6].
Many causative factors for CPHP have been hypothesised in the literature and are commonly characterised as intrinsic or extrinsic. Intrinsic factors are characteristics of an individual that predispose them to injury [
6]. Those suggested in the literature include limited first metatarsophalangeal joint (MPJ) range of motion (ROM), limited ankle dorsiflexion ROM, leg length discrepancy, reduced heel pad thickness, increased plantar fascia thickness, excessive foot pronation, reduced calf strength, calcaneal spur, older age and increased body mass index (BMI) [
1,
7,
8]. Environmental and circumstantial influences acting upon an individual are known as extrinsic factors, and include prolonged standing, inappropriate shoe fit, previous injury and running surface, speed, frequency and distance per week [
1,
6,
7]. Empirical evidence for most of these factors is limited or absent [
9], meaning that the role (if any) of each of these factors in the development of CPHP is poorly understood.
In an attempt to help address this lack of empirical evidence, a matched case-control study was undertaken to examine the association between CPHP and a number of causative factors suggested in the literature. Factors for inclusion into the study were selected because they each had a small amount of evidence supporting an association with CPHP [
9], which required further investigation. As it was obviously impractical to examine all factors requiring further investigation, the authors attempted to select those factors that are routinely assessed by clinicians in the management of heel pain. It was hypothesised that pronated foot posture, increased BMI, decreased ankle dorsiflexion ROM, increased occupational lower limb stress and decreased calf endurance would all be associated with CPHP.
Discussion
The purpose of this study was to examine the association between CPHP and a number of commonly hypothesised causative factors. Univariate analyses showed an association between CPHP and increased BMI, pronated foot posture and increased ankle dorsiflexion ROM, whilst occupational lower limb stress and calf endurance showed no association. Multivariate logistic regression showed that those with CPHP were 3.7 times more likely to have a pronated foot posture (FPI ≥ 4) and 2.9 times more likely to be obese (BMI ≥ 30 kg/m2).
The association found between a pronated foot type and CPHP is supported by research indicating that increased strain is placed on the plantar fascia when the foot is placed in a pronated position [
19‐
21]. It is important to note that while foot posture has been shown to alter slightly over the course of a lifetime, the change is so slow that it essentially remains constant from one decade to the next [
22]. This means that although causality cannot be established in case-control studies, the foot posture of the case group participants is unlikely to have altered after the onset of the condition, and therefore pronated foot posture may also be a risk factor for CPHP.
The association found between increased BMI and CPHP is also supported in the literature, with four of the five previous studies to examine BMI in a non-athletic population also finding an association with increased BMI [
9]. The control group appeared to be representative of the wider population, as the proportion of the group found to be obese (BMI ≥ 30 kg/m
2) was the same value as for the Australian population in the 45–54 year age bracket (21%) [
23]. Due to the fact that this study cannot establish causality, it is unclear whether increased BMI existed in the case group participants prior to the development of CPHP, or whether the pain associated with the condition caused participants to reduce their physical activity, thereby leading to an increase in BMI. However, it is plausible that increased BMI may be a risk factor for CPHP, as individuals with increased BMI experience higher vertical forces under the heel during gait [
24], leading to higher internal stresses within the heel [
25], which may lead to damage of soft tissue structures and the development of symptoms.
The identified association between increased ankle dorsiflexion ROM and CPHP was contrary to the common clinical perspective that decreased ankle dorsiflexion ROM is a causative factor for CPHP. This hypothesis is based on the theory that equinus (ankle dorsiflexion less than 10°) during gait causes abnormal compensatory pronation of the subtalar joint, which in turn increases stress on the plantar fascia [
26]. Although this theory is widely accepted, research evidence to support it is weak [
27]. Cornwall and McPoil [
27] found that a mild-to-moderate loss of passive dorsiflexion ROM (the study group had a mean ROM of 9.6°) had little or no effect on the frontal plane function of the rearfoot during the stance phase of gait. The participants were found to compensate for their dorsiflexion deficit with alterations in gait timing. This may explain why only one of three previous case-control studies found an association between decreased ankle dorsiflexion ROM and CPHP [
9].
A possible explanation for the identified association with increased ankle dorsiflexion ROM is that a non-linear relationship may exist between ankle dorsiflexion ROM and plantar fascia strain. If the relationship were U-shaped, both extremes of movement (increased and decreased ROM) would predispose to CPHP. To further substantiate this hypothesis, research would be required to determine whether increased translation of the tibia over the foot increases strain on the plantar fascia.
A criticism of the Dorsiflexion Lunge Test is that the test procedure makes no effort to control for pronation or supination of the foot. As increased subtalar joint pronation is known to increase the amount of dorsiflexion that can occur at the midtarsal joint, it is plausible that the increased ankle dorsiflexion ROM observed in the case group may have been due to the fact that the group was also found to have a more pronated foot posture [
28]. However, chi-square analysis indicated that scores for the Dorsiflexion Lunge Test and FPI were not correlated (data not shown), indicating that dorsiflexion was likely to be increased independent of foot posture in the case group participants.
A number of differences exist between the current study and previous literature with regard to dorsiflexion testing methods and results. All previous studies to address the association between ankle dorsiflexion and CPHP examined subjects with the knee extended [
9]. Since knee extension biases the gastrocnemius muscle whilst knee flexion has a Soleus bias, it is possible that tightness in the gastrocnemius muscle may have gone undetected in the case group. Also, the Dorsiflexion Lunge Test scores reported in the current study are slightly lower than those documented in the literature [
14,
29]. Because ankle dorsiflexion ROM decreases with age [
30], this difference is likely to be due to the fact that the mean participant age in the current study was approximately 30 years greater than in these previous studies. In summary, the findings of the current study question the role of decreased ankle dorsiflexion ROM in the development of CPHP and suggest increased ankle dorsiflexion ROM as a previously unconsidered causative factor.
Prolonged standing is often cited as a causative factor for CPHP [
1,
6,
7], based on the theory that prolonged tensile loading of the plantar fascia predisposes individuals to the condition [
31]. There is a weak level of evidence to support an association between prolonged standing and CPHP [
9], however, no previous study has adequately defined prolonged standing. Consequently, there are no data to indicate what activities are commonly performed whilst standing and therefore the nature of the stresses placed on the lower limb [
9]. This study was the first to examine prolonged standing in detail, using the Occupational Rating Scale to quantify the stresses placed on the lower limb during an average working day.
As there was no significant difference in the presence co-morbidities between the case and control groups (Table
1), any differences observed between the groups on the Occupational Rating Scale can be cautiously attributed to CPHP. However, no association was found between CPHP and average time spent sitting, standing, walking on uneven ground, squatting, climbing, lifting/carrying or total stress placed on the lower limb. The one significant difference found between the groups indicated an association between CPHP and reduced weight carried. Although this association was identified, it is unlikely that this factor has a role in the development of CPHP. It is more likely that the pain associated with CPHP causes sufferers to carry less weight than they otherwise would.
Although no association was identified with CPHP, it is unclear from these results whether occupational lower limb stress plays a role in the development of the condition. Due to the case-control design of the current study, it is possible that the case group participants experienced higher occupational lower limb stress prior to developing CPHP. The participants may have simply reduced their activity levels, as a consequence of their pain, to a level comparable to the control group. As participants were asked to answer the Occupational Rating Scale according to their current work status, no retrospective comments can be made regarding the association between CPHP and past working history. However, keeping this limitation in mind, it can be cautiously speculated that greater occupational lower limb stress may not be a risk factor for CPHP as previously thought.
No association was identified between calf endurance and CPHP. The Standing Heel Rise Test scores reported for the case and control groups were substantially lower than those reported in the literature, however, this was to be expected. Ross and Fontenot [
17] examined a far younger (21.2 ± 1.3 years) and more physically active sample of air force cadets and Lunsford and Perry [
32] examined a younger (male: 34.7 ± 8.5; female: 29.3 ± 5.0) sample using a test procedure that allowed a reduction of up to 50% in plantarflexion ROM before termination. As with occupational lower limb stress, it can be speculated from these findings that decreased calf endurance may not play a role in the development of CPHP.
The findings of this study need to be interpreted in light of a number of study limitations. A different investigator was used for each group and no pilot study was conducted to examine the correlation between the investigators for any of the outcome measures used. The Dorsiflexion Lunge Test and Standing Heel Rise Test have structured protocols and the Dorsiflexion Lunge Test has demonstrated high inter-tester reliability in previous studies [
14]. It is therefore unlikely that a change in investigator would have substantially altered the results of these measures. The FPI protocol involves a degree of subjectivity due to its observational nature and a change in examiner may have influenced the results. However, the authors are confident that the procedure is reliable enough between examiners to dichotomise participants as having pronated feet or not. Finally, as previously acknowledged, a case-control study cannot imply causation. As such, further research is required to definitively establish whether the associated factors identified in this paper are in fact risk factors for CPHP.
A representative clinical population with heel pain was used in this study; participants were included according to clinical signs and symptoms (chronic plantar heel pain) rather than diagnostic imaging. As CPHP can include a range of pathologies affecting the heel including plantar fasciitis, sub calcaneal bursitis, calcaneal periostitis and subcalcaneal spur, a pain diagnosis was considered the most appropriate method of selecting participants. These conditions can exhibit a combination of osseous and soft tissue pathologies (including calcaneal spurs, plantar fascial thickening, cortical irregularities and fat pad abnormalities [
4]) that have variable imaging findings [
33‐
35]. The sample would have been reduced to a specific sub-group of people with CPHP if a single imaging modality had been used as the inclusion criteria. Furthermore, diagnostic imaging is not always necessary for the diagnosis of CPHP, and many health professionals who frequently treat the condition (such as podiatrists and physiotherapists) rely on clinical criteria. We therefore believe that the use of a clinical diagnosis for inclusion into the CPHP group provides results that can be generalised to the broader population of people seeking treatment for heel pain.
A final limitation is that the overall classification accuracy of the model was relatively low (66% of cases correctly classified), which indicates that there may be other variables of importance that were not included in our test battery. Further research is required to determine whether the inclusion of other postulated risk factors can improve the classification accuracy of the multivariate model.
Competing interests
The author(s) declare that they have no competing interests.
Authors' contributions
DBI participated in the design of the study, was involved in data collection and prepared the manuscript. JLC conceived of the study, participated in its design and coordination and helped to draft the manuscript. MAY was involved in data collection and drafting of the manuscript and HBM participated in the design of the study, performed the statistical analysis and helped to draft the manuscript. All authors read and approved the final manuscript.