Introduction

Heel pain is a clinical condition that is quite common in almost every decade of life. Due to various differential diagnosis, correct treatment is hard to decide. Haglund’s syndrome (HS) is the most common reason of heel pain. It was first described in 1927 as retrocalcaneal bursitis caused by posterior calcaneus’s superior bony prominence (1). Over time, painful heel conditions such as; retrocalcaneal bursitis, supracalcaneal bursitis, and Achilles tendinitis were also defined as HS. Although the cause is unknown; etiological factors, including wearing tight shoes and repetitive trauma, are blamed. To understand and predict HS, several different measurement techniques have been reported in the literature, unfortunately none of these methods have been consistent or reliable to be  associated with Haglund’s deformities (2).   

This retrospective study aims to compare the radiological parameters of the normal population with the patients diagnosed as having HS to measure the reliability of the radiological parameters in the diagnosis of Haglund’s disease. We hypothesized that increased abnormality of the radiological parameters would associate with more pain and decreased functionality in patients with HS.

Methods

Patients

After acquiring ethical committee approval, 31 patients treated with the diagnosis of HS over a 5-year period between January 2014 and September 2019, and 36 healthy subjects without hindfoot pathologies who were seen at the consultation for other foot pathologies, were included in the study. Informed consent was signed by all of the included patients. The Haglund group consisted of 17 female and 14 male patients, with a mean age of 47.61 years (range, 31-68 years), with 16 right and 15 left feet. Inclusion criteria comprised calcaneal insertion pain on palpation of the posterior tuberosity and distal calcaneal tendon insertion, calcaneal hump, irritation of adjacent skin by shoe wear back-strip, painful pre or retro-calcaneal bursitis, and radiologic enthesophytes. The control group consisted of 24 female and 12 male patients, with a mean age of 44.7 years (range, 18-68 years), with 17 right and 19 left feet. These patients had no posterior heel pain and hindfoot pathology or radiographic enthesophytes in the calcaneal insertion region. The exclusion criteria were as follows; history of foot surgery, foot implant, amputation of part of the foot, fractures, injury requiring cast or surgery, calcaneal tendon body tendinopathy, posterior impingement, osteochondral defect, neuromuscular pathology, diagnosed as having muscle or skeletal abnormalities, inflammatory rheumatism, osteoarthritis, congenital foot deformity, paralysis or current pregnancy. The exclusion criteria were the same in both groups. Visual analog scale (VAS) and AOFAS scores were recorded only for patients in the Haglund group.

The study was approved by the local ethics committee (date 16/07/2020 and decision number 2020-10) and was conducted in line with the principles of the Declaration of Helsinki.

Imaging 

A standard lateral radiograph of the weight-bearing foot, also including ankle, was taken. All radiographs were available digitally in our database. The software available in our system was radiographically able to measure angles with a precision of two decimal points. The following angles were recorded from the lateral radiographs: Fowler Philip angle (FPA), lateral talus-first metatarsal angle (LTFMA, Meary’s angle), and calcaneal pitch angle (CPA).

The FPA is subtended by a tangent to the posterior edge of the greater tuberosity of the calcaneus and a line between the lowest weight-bearing point of the posteromedial tuberosity and the end of the calcaneocuboid joint line (Figure 1). Normal values range between 44° and 69°, and the values >75° are considered pathological (3). CPA is the angle determined by the intersection of the baseline tangent to the anterior tubercle and the medial tuberosity with the horizontal surface (Figure 2). The normal range is 17-32 (4). LTFMA is the angle created between the bisection of the first metatarsal and a line perpendicular to a line connecting the anterodorsal and anterior-plantar extremes of the talar head (Figure 3). The positive value indicates the talus is plantarflexed according to the first metatarsal. The negative value indicates pes planus (5).

To determine interobserver reliability, radiologic measurements were made by 2 independent observers (orthopedic surgeons). To assess intra-observer reliability, measurements were made twice by each observer at a minimum 14-days interval. Measurements were made independently by the two observers, blind to the patient’s group, known only to the principal investigator.

Statistical Analysis

The data obtained in this study were analyzed with IBM Statistical Package for Social Sciences (SPSS) Statistics 26 software (SPSS Inc., Chicago, IL, USA). The normality of the distribution for the variables was tested with the Kolmogrov-Smirnov test. The comparison of the means’ values from two variables with normal distribution was performed with Independent Samples t-test. To evaluate intra-and inter-observer reliability, Pearson correlation analysis was used according to the normality test results. In the evaluation of the correlation coefficient, r:=0-0.24 was considered as poor, r=0.25-0.49 as moderate, r=0.50-0.74 as strong, and r=0.75-1.0 as very strong. Quantitative data were expressed as mean, standard deviation, minimum and maximum values. The confidence interval was 95%, a p-value less than 0.05 was considered statistically significant.

Results

Mean CPA, FPA, and LTFMA in the Haglund group were 23.88±4.6, 58.16±5.7, and 4.92±1.88, respectively, and the same recordings in the control group were 21.16±4.81, 59.1±4.3, and 4.25±2.57 respectively. There was no significant difference between the Haglund and control groups in terms of FPA and LTFMA (p=0.47, p=0.22). However, CPA was found significantly higher in the Haglund group (p=0.021). All the results are summarized in Table 1.

Gender and age showed no correlation with the radiological parameters (CPA, FPA, and LTFMA)  in both groups (Table 2).

The average values of the VAS and AOFAS scores of the patients in the Haglund group were 8.45±1.06 and 47.4±7.58, respectively. Statistical evaluation of the patients’ VAS and AOFAS scores in Haglund group  showed no correlation with CPA, FPA, and LTFMA measurements. Also, there was no significant difference in terms of the AOFAS and VAS scores of the patients in the Haglund group between genders (p=0.393 and p=0.509, respectively) (Table 3).

Discussion

HS is one of the many causes of chronic posterior heel pain and is usually characterized by a posterosuperior bony prominence of the calcaneus, which is also defined as Haglund’s deformity (6). There are various radiological parameters used to measure the bursal projection of calcaneus in the literature, such as; Fowler Philip angle, Angle of Steffensen and Evensen, and the parallel pitch lines (7). However, the relationship between any of these measurement techniques and symptomatic posterior heel pain is unclear. FPA angle (> 75 degrees), with a “prow-like” projection of the posterosuperior prominence of the calcaneus and painful swelling of the soft tissues surrounding the Achilles tendon insertion, were found to be related (3). However, it was also reported that the CPA or vertical inclination of the calcaneus might be more theoretically relevant than FPA in the HS (8).

Lu et al. (9) and Fiamengo et al. (10) found that the radiographic measurement techniques associated with heel pain were inconsistent and not reliable enough to assist in the decision making for surgery. Heneghan and Pavlov (11) also observed 100% false-negative results using FPA. Ruch (12) and, Fugslang, and Torup (13) observed 86 to 88% false-negative results while using FPA. Also, Fugslang and Torup (13)  observed wide range of FPA variations in the non-Haglund group. Similarly, in other studies, FPA measurement below 75 degrees was detected at rates ranging from 86% to 100%, and high false-negative ratios were reported (9,11-13).

Our study analyzed and compared the radiographic measurements of LTFMA, CPA, and FPA of  patients with HS with the normal population. FPA, LTFMA, and CPA are considered to be indicators of foot shape (14). We hypothesized that the presence of high arch or cavus foot will irritate Achilles tendon due to the positioning of the calcaneus and exacerbate HS. According to our hypothesis, CPA and FPA will be used to detect calcaneal positioning and LTFMA will be used for detecting cavus foot. An increase in CPA and FPA would cause the calcaneus’ posterosuperior tubercle to become more prominent and thus, the Achilles tendon and retro-calcaneal bursa would be more irritated. According to our current knowledge, no previous study had reported comparison of both CPA, FPA and LTFMA in the normal population with the patients with HS . 

In our study, there was no significant difference between the Haglund group (58.16±5.7) and the control group (59.1±4.3) in terms of FPA (p=0.47). In the Haglund group, there was no patient with FPA more than 75 degrees, and 100% false-negative results were observed using FPA in our study so that our study was compatible with the findings of the studies of Fiamengo et al. (10) and Pavlov et al. (8) in this respect.

Singh et al. (15) concluded that CPA was increased in 57% of the control group and 63% of patients with Haglund’s deformity. Bulsara et al. (4) reported that CPA was significantly (p=0.021) higher in the Haglund group (22.14±4.74) than in the control group (20.28±5.12). Similarly, in our study, CPA was higher in the Haglund group (23.69±4.64) than in the control group (21.16±4.81). Although this value was within the normal range for CPA, it was higher than the normal control group, and this difference was statistically significant (p=0.021). Our study is compatible with the findings of Bulstra et al. (4) and Singh et al. (15) in this respect. 

The LTFMA > + 4^° is an indication of the cavus foot (6). In our study, the average LTFMA of the Haglund group was higher than the patients in the normal population (4.92±1.88 and 4.25±2.57, respectively). Although there is no statistically significant difference (p=0.22), this situation, together with increased CPA, shows that the arch height of the patients’ feet in the Haglund group is higher than in the control group, which can be attributed as a predisposing factor for the HS. The LTFMA in the normal population was reported as 4±5.5 in the study of Thomas et al. (16) and 5.5±3.9 in the study of Lamm et al. (17). Ahn et al. (6) reported that the lateral LTFMA and CPA were higher in patients with symptomatic HS. Researchers measured LTFMA as 5.9±5.0 preoperatively, and 6.3±4.7 postoperatively in patients with HS, and they reported that postoperative LTFMA was increased, but this increase was not statistically significant. However, only patients with symptomatic HS were included in this study, and patients in the normal population were not evaluated. Unlike the other studies, radiological parameters were measured in normal and Haglund groups and were compared in our study. Our findings have shown that LTFMA is not suitable for the assessment of the HS because of the difficulty in the measurement of LTFMA as it is in a narrow range, but increased CPA may be an indicator of HS.

Study Limitations

Our study’s reliability among observers was 0.902 for FPA, 0.874 for CPA, and 0.662 for LTFMA. The observers’ reliability was higher between the observers for FPA and CPA, and this was similar to the previous studies. (4,9,18) Although reliability within the observer increases with observer experience, we believe that the selected measurements should be easy to teach to acquire higher reliability (18).

Conclusion

Our results showed that LTFMA and CPA were higher in patients with symptomatic HS than in the normal population. This was statistically significant in CPA but not in LTFMA, but there was no correlation between the increment of CPA in regard of pain and functional status (VAS and AOFAS) of the Haglund group. We think that the measurement of CPA is more important in patients with heel pain. Bursal projection in a cavus foot is more prominent due to a more vertical calcaneus, even with a normal FPA. Increasing the vertical angle of the calcaneus increases the dorsal prominence of the bone and creates an important pathogenic factor. This situation can lead to an inflammatory process caused by relatively increased traction in the tendon. In other words, a dorsiflexed calcaneus may cause heel pain and HS.

The study’s retrospective nature, the relatively small number of patients involved and unavailability of MRI were limitations of the study. Studies with a larger number of patients using more adequate radiographic tools such as 3D CT scans may give more insight into the increase in the vertical angle of the calcaneus, the relationship between HS and cavus foot deformity, and the pathological formation in the calcaneus. Also, MRI could be useful to exclude other pathologies that might cause heel pain to homogenize the population.

In conclusion, in everyday practice of an orthopedic foot and ankle surgeon, Haglund’s deformity is an important part of heel pain complaints; although the posterolateral calcaneus’s swelling is seen on physical examination, it is not possible to distinguish the orientation of the calcaneus. In our study, HS was found to be associated with a higher CPA, but the pain intensity and functionality were not correlated with the increment of CPA. Although the clinical symptoms are the main indicators of the HS, radiological measurements especially showing the cavus deformity should be taken into account as an auxiliary tool to start the medication or decide on surgery.

Ethics

Ethics Committee Approval: The study was approved by the local ethics committee (decision number: 2020-10 and date:16/07/2020) and was conducted in line with the principles of the Declaration of Helsinki.

Peer-review: Externally peer reviewed.

Authorship Contributions

Surgical and Medical Practices: K.B., B.K., G.A., M.U.M., Concept:  K.B., L.A., B.K., G.A., M.U.M., Design:   B.K., O.Y., G.A., M.U.M., Data Collection or Processing:  K.B., L.A.,  B.K., O.Y., G.A., Analysis or Interpretation:  L.A., O.Y., Literature Search:  K.B., L.A., B.K.,  Writing:  K.B., L.A., B.K.,  M.U.M.

Conflict of Interest: The authors declared no conflicts of interest.

Financial Disclosure: The authors declared that this study received no financial support.