Given the lack of studies that have specifically addressed hallux rigidus in runners, this author reviews the existing literature on the condition, defuses a couple of biomechanical myths and offers salient pointers on a variety of treatment options ranging from orthotic therapy to arthrodesis.
Perhaps no other condition treated by the podiatric physician is more controversial and more poorly understood than hallux rigidus. Even less understood are the effects of this pathology on the running athlete or the selection of proper treatment interventions that ensure optimal return to athletic performance.
Accordingly, let us take a closer look at several myths and misunderstandings about the evaluation and treatment of hallux rigidus in running athletes.
For this article, the definition of a true “athlete” is any individual who runs more than 20 miles per week or participates in vigorous competitive sports such as tennis, racquetball, volleyball or basketball more than two times per week. Running athletes who present with hallux rigidus pose a significant challenge for the treating physician based upon the poor understanding of the biomechanics of this condition as well as a lack of agreement for a standardized treatment protocol.
While it may be universally accepted that one should initially pursue conservative, non-operative treatment when treating an athlete with hallux rigidus, there may be a lack of consensus on the type of conservative interventions. If you use foot orthotic therapy, what are the casting and prescription criteria? Are corticosteroid injections indicated? What is the role of physical therapy? Are footwear modifications indicated? 
Finally, when conservative measures fail, what are the best surgical options for the running athlete? Are there published studies that document favorable outcomes for the running athlete with hallux rigidus?
Cotterill first coined the term hallux rigidus in 1887.1 Hallux rigidus is now the most universally accepted description of a condition in which there is a combination of restricted range of motion and degenerative arthrosis of the first metatarsophalangeal joint (MPJ). The term “functional hallux limitus,” originally described by Laird, should apply to a separate group of individuals who do not have degenerative changes of their great toe joints.2 However, it is recognized that functional hallux limitus may be one of the etiologic factors of hallux rigidus.
Various authors have described hallux rigidus as resulting from a myriad of causes including systemic arthrosis, trauma, inflammatory disorders, neuromuscular disorders, congenital abnormality and iatrogenic events. The prevailing thinking is that abnormality in dynamic foot function is the primary etiology of hallux rigidus. However, the true mechanism of dysfunction of the great toe joint during gait remains poorly understood.
Roukis has written the most insightful and provocative papers on hallux rigidus in the podiatric literature.3-5 In these studies, Roukis demonstrated a clear relationship between hallux equinus and metatarsus primus elevatus in patients with hallux rigidus. What has not been demonstrated is whether these clinical and radiographic measurements were acquired or congenital, or what biomechanical conditions were responsible for the progressive changes in alignment on the medial column of the foot. 
There have been many classification systems relevant to hallux rigidus. In a recently published in-depth analysis of these systems, Beeson et al., demonstrated universal shortcomings and the need for valid and reliable criteria.6 Without reliable, objective measures of range of motion of the first MPJ joint and sound criteria for the measurement of degenerative changes using traditional imaging techniques, it will not be possible to develop treatment protocols that can have reliable outcomes.
Notwithstanding, Roukis, et al., have proposed a classification system, which is actually a hybrid of the systems proposed by Drago, Hanft and Kravitz.3,7-9 This system appears to be most useful in the evaluation of the patient with hallux rigidus. (See “A Guide To Radiographic Grading Of Hallux Rigidus” on page 54.) In terms of running athletes, there is no existing classification system that takes into account the level of activity of the patient, the amount of pain and disability during exercise, or the footwear requirements of the sport. For example, moderate osteophytic spurring may be asymptomatic in the sedentary patient wearing roomy oxford shoes but this can become severely painful for an athlete wearing soccer cleats.
Most classification systems for hallux rigidus do not provide objective criteria for measuring range of motion of the first MPJ joint. Furthermore, there is significant misunderstanding of what is considered “normal” range of motion of the great toe joint, whether one is assessing this with the patient in a non-weightbearing position or during walking and running gait.
Nearly every podiatric physician assumes the normal dorsiflexion range of motion of the first MPJ is at least 65 degrees. One reason is the universally accepted proposal by Root, Weed and Orien, who demonstrated in a theoretical model that 65 degrees of dorsiflexion demand would be required of the hallux on the first metatarsal if normal hip, knee and ankle flexion occurred during the propulsive phase of gait.10 
Since the work of Root, et al., was published, many other authors (Joseph, Buell, Bojsen-Moller, Hetherington) speculated or measured “normal” ranges of motion of the first MPJ, which ranged between 65 and 100 degrees of dorsiflexion.11-13 Further scrutiny reveals that many of these studies confuse non-weightbearing measurements of range of motion of the great toe joint with actual measurements obtained during gait. In reality, until recently, no accurate methodology existed to measure motion of the first MPJ in gait.
Until recently, measuring range of motion of bone segments within the foot was not possible. Today, studies of measurements of multi-segment foot models are being published regularly and represent the most exciting research of lower extremity function. These studies have already challenged many previously held notions about the movement of bones in the human foot during gait.
In 1999, Nawoczenski, et al., published a study of motion of the first MPJ in 10 healthy people during walking gait.14 They used an electromagnetic tracking device to accurately measure motion of the hallux relative to the first metatarsal during the entire stance phase of gait.
Nawoczenski, et al., also studied different clinical tests to determine which measure would most accurately predict the actual range of motion patients would utilize in their great toe joint during gait. These tests included: passive dorsiflexion (non-weightbearing); active dorsiflexion (non-weightbearing); passive dorsiflexion (weightbearing); active dorsiflexion (weightbearing); and active dorsiflexion during heel raise.
In ten healthy people, the average dorsiflexion range of motion of the first MPJ during walking gait was 42 degrees. The average passive dorsiflexion range of motion with non-weightbearing (most commonly used by podiatric physicians) was 57 degrees. This measurement overestimated the actual range of motion during gait. The active dorsiflexion with weightbearing was 44 degrees and most accurately predicted the dorsiflexion range used during gait by the study participants. 
This study by Nawoczenski dispels previous notions about clinical evaluation of range of motion of the first MPJ. Furthermore, a new standard of “normal” range of dorsiflexion range of motion of the great toe joint should now be set at approximately 45 degrees. However, this dorsiflexion range has only been verified for walking gait, not running.
The assumption that running amplifies everything more than walking is somewhat true. Impact forces, muscular activation and joint moments are just a few factors that are substantially greater during running in comparison to walking. However, the assumption that all of the joints of the lower extremity go through greater range of motion during running, in comparison to walking, is simply not true.
Early on in my career, I became aware that many of my runner patients with hallux rigidus would report less symptoms during running than walking. A review of the literature has shown no valid study of measurement of range of the motion of the first MPJ during running gait. However, conversations with many respected podiatric physicians confirmed a belief that running gait would require greater range of motion of the great toe joint than would walking gait. Accordingly, one would expect that hallux rigidus symptoms would become worse during running. My experience has been just the opposite.
Mari Adad, DPM, has provided further insight with unpublished data from a study of six healthy individuals. Using the same type of electromagnetic tracking system utilized by Nawoczenski to measure range of motion of the first MPJ, Adad measured an average of 34 degrees dorsiflexion of the great toe joint during walking and only an average of 26 degrees dorsiflexion during running. 
In regard to understanding the reduced dorsiflexion range of motion of the first MPJ in running in comparison to walking, Sasaki and Neptune used computer modeling to predict joint angles and forces of the lower extremity segments during running.15 This study confirmed that ground reaction forces peak earlier under the forefoot during running in comparison to walking when the ankle is in a less plantarflexed position. A reduced ankle plantarflexion angle during running, in comparison to walking, puts less dorsiflexion demand on the first MPJ during the heel rise portion of the gait cycle.
The earlier and greater peak ground reaction force in running is partly due to increased muscular activation of the ankle flexors including the flexor hallucis longus. Early and greater activity of the flexor hallucis longus will restrict dorsiflexion of the first MPJ during running in comparison to walking. In other words, runners with hallux rigidus may use greater muscle activity to restrict painful dorsiflexion of the first MPJ and minimize symptoms in comparison to walking.
When an athlete initially presents with hallux rigidus, the symptoms are usually significant as this type of patient will put off a visit to the doctor as long as possible. It is suggested that treatment of hallux rigidus on the initial visit be similar to treating any other acute injury in the athlete. One should follow the protocol of PRICE (protection, rest, ice, compression and elevation) to calm down initial symptoms. Often, a brief period of rest from running activities while substituting non-impact cardiovascular training (i.e. bike, elliptical trainer) can allow acute inflammation and swelling to subside so follow-up conservative measures have a better chance of succeeding.
As I noted previously, many running athletes can participate in their sport with significant clinical evidence of hallux rigidus but have minimal symptoms. By the time they present to the podiatric physician for treatment, their condition may have already advanced to stage III or stage IV. At this point, degenerative changes in the first MPJ are so severe that most non-operative measures will have little hope of success.
However, bear in mind that running athletes use a smaller range of motion than walkers so small increments of a change of a range of motion may have a more profound benefit in relieving symptoms. The symptoms of hallux rigidus basically derive from either the degenerative arthrosis process, the mechanical jamming of the first MPJ, osteophytic impingement against surrounding soft tissue or footwear. Conservative treatment through the use of foot orthoses, physical therapy or footwear modification can address one or all of these causes of symptoms.
In treating the running athlete with custom functional foot orthoses, one can direct strategies toward one of two opposing functions. Physicians can either improve first ray function and dorsiflexion range of the first MPJ, or block range of motion of the hallux on the first metatarsal.
In stage I hallux rigidus, a standard orthotic prescription designed to improve subtalar position, locking of the midtarsal joint and stabilization of the first ray during propulsion can be very successful in minimizing symptoms. The foot orthosis casting should utilize a neutral suspension cast, holding the midtarsal joint loaded and fully pronated. Plantarflexing the first ray to end range will maximize position of this segment for optimal first MPJ dorsiflexion.
When it comes to enhancing plantarflexion of the first metatarsal, it may be beneficial to add to the orthosis with a reverse Morton’s extension or a Kinetic Wedge®. I have also been impressed with the new Cluffy Wedge®, which preloads the plantar fascia to facilitate early engagement of the windlass mechanism around the first MPJ.
When treating stage II through IV hallux rigidus in runners, the primary goal of foot orthotic therapy or shoe modification should be blocking or shielding the hallux from dorsiflexion at the first metatarsal. An extension of the footplate of the orthotic under the hallux is a simple yet very effective technique to reduce dorsiflexion moment at the first MPJ. The main drawback to this type of device is the bulk of the orthosis plate under the great toe joint as this can compromise shoe fit. If a patient has a prominent exostosis of the first MPJ, the increased bulk of the footplate extension may actually exacerbate symptoms rather than provide relief.
An alternative to foot orthotic footplate extensions under the hallux to block motion of the first MPJ is the use of a graphite composite plate inside of the shoe. These plates are available at various degrees of thickness and rigidity, and will fit in a cleated shoe better than a foot orthotic. The response of patients with hallux rigidus to blocking footplates is variable. For the most part, athletic performance is not hampered by the use of these devices.
The selection of footwear that has maximal stiffness across the forefoot will decrease the dorsiflexion moment directed across the MPJs during running gait. Most running shoes have some degree of flexibility across the forefoot. Therefore, I often direct runners with stage II and greater hallux rigidus to switch to lightweight day hikers and switch from asphalt to dirt trails for long distance running. Yes, these shoes may be too stiff for the average runner but they can facilitate profound relief of jamming of the great toe joint.
I do not advocate the use of injectable corticosteroids for athletes with hallux rigidus unless there may be some benefit for short-term relief of symptoms. For the most part, the relief is so temporary that this type of treatment is disappointing. Non-steroidal anti-inflammatories can relieve pain but do not reverse the pathology of hallux rigidus. Pain relief with medication may only mask the degenerative process and could lead to rapid deterioration of the great toe joint in running athletes.
Physical therapy offers a reasonable level of hope to the athlete with hallux rigidus who wants to avoid surgery. Just as we have learned with cartilage injuries to the knee, aggressive rehabilitation protocols using muscle reeducation, joint mobilization and flexibility exercises can stabilize a joint and minimize symptoms of degenerative arthritis. Strengthening of the flexor hallucis longus muscle as well as the plantar intrinsic muscles of the feet can improve stability of the first MPJ. A skilled therapist can also mobilize this joint to increase range of motion. All of these measures can be of value to the patient even if he or she ultimately undergoes surgery.
Surgery designed to relieve the symptoms of hallux rigidus can have one or more intended methods and outcomes. Surgeons may restore articular cartilage via stimulation of local chondrogenic source, transplant autograft tissue. One may perform decompression or realignment of the first MPJ via osteotomy of the first metatarsal or proximal phalanx. Alternately, surgeons may resect the joint surface(s) via arthroplasty with soft tissue interposition or a prosthetic implant, or perform arthrodesis of the first MPJ.
There has been a variety of papers over the last 40 years that have documented positive and negative outcomes of surgical procedures for the treatment of hallux rigidus. Few have focused on the long-term effects of any surgical treatment for running athletes. The majority of patients included in large group studies receiving surgery for hallux rigidus are over the age of 50, overweight and have a sedentary lifestyle. The positive outcome of many surgical procedures performed on these types of patients could not be expected in an athletic patient population.
With any surgical intervention for hallux rigidus, surgeons should perform an intraoperative inspection and debridement before making any decision about osteotomy, fusion, etc.
Many times, a patient showing significant symptoms of hallux rigidus with minimal radiographic evidence of degenerative joint disease will demonstrate unexpected cartilage loss when one surgically explores the joint. At other times, a rigid joint will suddenly demonstrate over 50 degrees dorsiflexion after simple resection of the osteophytes and freeing of the sesamoid apparatus adhering to the plantar surface of the first metatarsal.
One should perform subchondral drilling on all cartilage defects whether the surgeon opts for osteotomy or a simple cheilectomy. Newer techniques of autograft implantation of osteochondral material obtained from the talus offer promise for the treatment of early stage hallux rigidus. However, no studies have been published showing outcomes with running athletes.
One of the few papers documenting long-term outcomes of surgery on high level athletes with hallux rigidus utilized a cheilectomy procedure for relief of symptoms.16 The study authors noted good to excellent results after a five-year follow-up of 20 athletic patients and pedobarographic measurements showed improvements of pressure distribution under the forefoot with less lateral loading of the metatarsals.
The typical cheilectomy involves resection of one-third to one-fourth of the dorsal articular surface of the first metatarsal head. A modification of this involves a Valenti arthroplasty, which is a V-shaped resection of the dorsal half of the head of the first metatarsal and the base of the proximal phalanx. Papers advocating the use of an interpositional arthroplasty involving autograft or allograft tendon material have not studied results with running athletes.
It is important to restore range of motion to at least 50 degrees dorsiflexion intraoperatively. In my experience, approximately 10 to 20 degrees of dorsiflexion noted intraoperatively will be lost in the postoperative period. Therefore, if one desires an end result of 30 degrees dorsiflexion, the surgeon should obtain at least 50 degrees dorsiflexion on the operating table with any type of surgical procedure (other than arthrodesis) for hallux rigidus.
Recently, Nawoczenski, et al., published a detailed analysis of 20 patients who underwent cheilectomy for hallux rigidus and were surveyed six years after the procedure.17 The average dorsiflexion increased by 12 degrees postoperatively but the study patients did not achieve full range for normal walking. Yet the overall range of postoperative dorsiflexion was 30 degrees, which is perhaps considered “normal” for running activities.
Restoring alignment of the first MPJ would seem to be the most straightforward, sensible surgical solution for hallux rigidus. Unfortunately, the published results of these procedures have not validated predictable success in restoring alignment and have not focused on any groups of high performance athletes.
Roukis has evaluated reports of patients undergoing either a Green-Watermann osteotomy or a Youngswick-Austin osteotomy, and concluded there were no significant changes in metatarsus primus elevatus or hallux equinus after either osteotomy.5 Furthermore, patient satisfaction with joint interposition procedures or cheilectomy is similar to joint decompression procedures, according to Roukis, yet these procedures have much less chance for complication than osteotomies of the first metatarsal.
The primary concern with first metatarsal osteotomies is potential serious complications for the running athlete. Since these procedures are designed to shorten the first metatarsal and decompress the joint, the risk of transfer metatarsalgia is significant for all patients. For running athletes, whose peak forefoot forces are amplified threefold in comparison to when they walk, increased plantar pressure under the central metatarsals can lead to capsulitis, predislocation syndrome and stress fracture of the metatarsals.
When I perform an osteotomy for relief of hallux rigidus, I make certain that the first metatarsal is either longer than the second or that there is no significant increased pressure under the second metatarsal in comparison to a preoperative measurement with a Mat Scan® analysis. The best osteotomy to restore function of the first MPJ is the modified Hohman procedure, which can maximize plantarflexion with minimal shortening.18
Joint implants have been advocated for over 40 years for the treatment of hallux rigidus. However, there is currently no study documenting the long-term performance of any first MPJ prosthesis in running athletes. However, many manufacturers will advocate use of these implants for younger patients with active lifestyles that include vigorous running.
While runners may use less range of motion in the first MPJ than walkers, the magnitude of forces in running across this important joint cannot be overestimated. Until recently, it was not possible to measure joint moments within the foot during walking or running. Now, with multi-segment foot modeling, we will soon learn more about the forces transmitted within the medial column of the foot and the first MPJ.
For now, we can look at the ankle and compare the differences in forces that occur in walking versus running. Many investigators have studied forces or moments that occur at the ankle joint, and have found that there is an approximate 10-fold increase of internal joint moment during running as opposed to walking. This means that the internal structure of the joint (i.e. the articular cartilage) is subjected to a 10-fold increase of stress-strain during running activity.
The first MPJ is actually more complicated than the ankle since there is considerable frontal plane movement of the first metatarsal coupled with dorsiflexion of the hallux. No implant for the great toe joint is designed to allow this coupled movement in the same magnitude. Thus far, measurements have been in vivo and no implant has been tested under the vigorous conditions of running.
Given the lack of evidence of favorable outcomes with prosthetic joint implants for the treatment of hallux rigidus in running athletes, podiatrists should be cautious about using these devices in this patient population. Furthermore, when an implant fails, the options for salvage may end the career of the running athlete.
What was once a forbidden procedure for the active patient is now becoming the best option for the treatment of stage III and IV hallux rigidus. Out of all retrospective studies of surgical treatment of hallux rigidus, those utilizing arthrodesis of the first MPJ consistently show superior results and patient satisfaction in comparison to other surgical options.19-22
Concerns about fusion of the first MPJ in active individuals focus on alteration of gait mechanics, transfer of dorsiflexion moment to the hallux interphalangeal joint, loss of the windlass mechanism and an inability to wear elevated heel footwear. However, there is minimal evidence in the literature that any of these events actually occur to any significant level.
When it comes to studies of athletes, Bouché showed favorable results in his preliminary study published in 1996.23 Bouché has expanded his study to now include 40 active athletic patients who have achieved good to excellent results with arthrodesis of the first MPJ for painful hallux rigidus.24
Kinematic studies of patients who underwent arthrodesis of the first MPJ showed a significantly shorter step length as well as reduced ankle power and torque at push off.20 At the same time, Lombardi showed increased medial longitudinal arch height after fusion of the first MPJ.19
It is unclear whether the gait abnormalities are the result of fusion or are a long-term compensation of walking with a painful first MPJ that has not changed after surgery. The fact is that athletes who present for treatment of stage III and IV hallux rigidus have functioned with less than 10 degrees range of motion of the first MPJ for many years.
Fusion of this joint likely has little change on gait and compensation at this point. The primary effect of fusion is the elimination of pain, which has demonstrated significant positive patient satisfaction.
Running athletes with hallux rigidus are unique and differ from the typical older sedentary patients who most frequently present with this condition. On one hand, running athletes demand less dorsiflexion range of motion of the first MPJ in comparison to walking patients. Therefore, runners can continue their sport even when dorsiflexion range of motion diminishes to less than 40 degrees. Conversely, running increases internal joint moments in the feet 10-fold, which should be a strong deterrent to the use of prosthetic joint implants that have been designed to relieve the symptoms of hallux rigidus in sedentary patients.
Few studies have evaluated the effectiveness of surgical interventions for the treatment of hallux rigidus in running athletes. While cheilectomy may be beneficial for early stages of hallux rigidus, arthrodesis of the first MPJ appears to be the best option for the relief of symptoms with stage III and stage IV hallux rigidus in active, athletic patients.
Dr. Richie is an Adjunct Associate Professor in the Department of Applied Biomechanics at the California School of Podiatric Medicine at Samuel Merritt College. He is a Fellow of the American Academy of Podiatric Sports Medicine and is in private practice in Seal Beach, Ca.
For further reading, see “Managing Hallux Rigidus In The Athlete” in the April 2004 issue or “Key Insights In Treating Hallux Limitus” in the March 2007 issue of Podiatry Today.
For reprint information or to access the archives, visit www.podiatrytoday.com.
1. Cotterill JM. Condition of stiff great toe in adolescents. Edinburgh Med J 33: 459, 1887.
2. Laird PO. Functional hallux limitus. Illinois Podiatrist 9:4,1972.
3. Roukis TS, Jacobs PM, Dawson DM, et al. A prospective comparison of clinical, radiographic and intraoperative features of hallux rigidus. J Foot Ankle Surg 41(2): 76-95, 2002.
4. Roukis TS, Jacobs PM, Dawson DM et al. A prospective comparison of clinical, radiographic and intraoperative features of halux rigidus: short-term follow up and analysis. J Foot Ankle Surg 41(3): 158-65, 2002.
5. Roukis TS. Metatarsus primus elevatus in hallux rigidus fact or fiction? JAPMA 95(3): 221-8, 2005.
6. Beeson P, Phillips C, Corr S, et al. Classification systems for hallux rigidus: A review of the literature. Foot Ankle Int 29(4): 407-14, 2008.
7. Drago JJ, Oloff L, Jacobs AM. A comprehensive review of hallux limitus. J Foot Surg 23: 213, 1984
8. Hanft JR, Mason ET, Landsman AS, Kashuk KB. A new radiographic classification for hallux limits. J Foot Ankle Surg 32: 397, 1993
9. Kravitz SR, LaPorta GA, Lawton JH. Progressive staging classification of hallux limitus and hallux rigidus. Lower Extremity 1: 55, 1994.
10. Root ML, Orien WP, Weed JH. Normal and abnormal function of the foot. In Clincal Biomechanics, vol II, Clinical Biomechanics Corp., Los Angeles, 1977
11. Buell T, Green D, Risser J. Measurement of the first metatarsophalangeal joint range of motion. JAPMA 78:439, 1988.
12. Bojsen-Moller F, Lamoreux L. Significance of free dorsiflexion of the toes in walking. Acta Orthop Scand 50: 471, 1979
13. Hetherington VJ, Johnson RE, Albritton JS. Necessary dorsiflexion of the first metatarsophalangeal joint during gait. J Foot Surg 29:218, 1990.
14. Nawoczenski DA, Baumhauer JF, Umberger BR. Relationship between clinical measurements and motion of the first metatarsophangeal joint during gait. J Bone Joint Surg 81(3): 370-6, 1999.
15. Sasaki K, Neptune RR. Differences in muscle function in walking and running at the same speed. J Biomechanics 39(11): 2005-13, 2006.
16. Muller T, Steenwerckx A, Thienpont E. Results after cheilectomy in athletes with hallux rigidus. Foot Ankle Int 20(4): 232-7, 1999.
17. Nawoczenski DA, Ketz J, Baumhauer JF. Dynamic kinematic and plantar pressure changes following cheilectomy for hallux rigidus: A mid-term followup. Foot Ankle Int 29(3), 265-72, 2008.
18. Gonzalez JV, Garrett PP, Jordan MJ, Reilly CH. The modified Hohmann osteotomy: An alternative joint salvage procedure for hallux rigidus. J Foot Ankle Surg 43(6): 380-8, 2004.
19. Lombardi CM, Silhanek AD, Connolly FG, Dennis LN. The effect of first metarsophalangeal joint arthodesis for treatment of hallux rigidus: A retrospective study. J Foot Ankle Surg 40(3):137-43, 2001.
20. DeFrino PF, Brodsky JW, Pollo FE, Crenshaw SJ, Beischer AD. First metarsophalangeal arthrodesis: A clinical, pedobarographic and gait analysis study. Foot Ankle Int 23(6): 496-502, 2002.
21. Taylor DT, Sage RA, Pinzur MS. Arthrodesis of the first metatarsophalangeal joint. Am J Orthop 33(6): 285-8, 2004.
22. Gibson JN, Thomson CF. Arthrodesis or total replacement for hallux rigidus: A randomized controlled trial. Foot Ankle Int 26(9): 680-90, 2005.
23. Bouche RT, Adad JM. Arthrodesis of the first metatarsophalangeal joint in active people. Clin Pod Med Surg 13(3);461-84, 1996.
24. Personal communication with Richard Bouche, DPM.