@

( oE@\ ru.t@iI @)S4

J('urn(i/ f Ii(flt( ,poI JOt??!Surgen . 1n rporaud

Fractures of the Base of the Fifth Metatarsal Distal to the Tuberosity

CLASSIFICATION AND GUIDELINES BY JOSEPH S. TORG, M.D.*, HELENE PAVLOV, FOR NON-SURGICAL M.D.*, AND SURGICAL MANAGEMENT RUSSELL R. ZELKO, M.D.t, AND MARIANNE DAS, B.S.*, ITHACA, FREDERICK C. BALDUINI, M.D4, PHILADELPHIA,

NEW YORK, N.Y. THOMAS C. PEFF, M.D.*, PHILADELPHIA, PENNSYLVANIA

1ro@n the 1)epartinent of Ort/zopaedic Surgery. University of Pennsylvania Sc/zoo! of Medicine. ciiid the Department of University Health Sciences. Cornell Unitersitv, Ithaca

Philadelphia.

ABSTRACT:

Between

1973

and

1982

forty-six

frac

vatively,

but an active athlete with delayed union will

tures of the base of the fifth metatarsal, distal to the tuberosity, were treated and followed for a mean of forty months (range, six to 108 months). Roentgenographic criteria were used to define three types of fractures: acute fractures characterized by a narrow fracture line and absence of intramedullary sclerosis; those with delayed union, with widening of the fracture line and evidence of intramedullary scleross; and those with non-union and complete obliteration of the medullary canal by sclerotic bone.

benefitfrommedullary curettage andbone-grafting, as

will fractures that have progressed to symptomatic non union. Fractures of the proximal part of the fifth metatarsal can be separated into two types: those involving the tuber osity and those involving the proximal part of the diaphysis distal to the tuberosity. Recently it has been recognized that the latter group, Jones fractures, may be difficult to treat. Although reports in the literature have indicated the potential difficulties in the treatment of Jones fracturest246, prevailing guidelines for their management are ambiguous. Apparently the varied clinical and roentgenographic mani festations of these fractures have not been correlated with their response to treatment. In this paper we describe a classification of these fractures and a plan of treatment based on clinical and roentgenographic criteria that were devel oped to define acute fractures, delayed unions, and non unions. Material and Methods Between 1973 and 1982 two of us (J. S. T. and R. R.
Z.) treated forty-three patients with forty-six fractures of

Of the twenty-fiveacute fractures in this series,

fifteen were treated with a non-weight-bearing toe-to knee cast, and fourteen of them healed in a mean of seven weeks. Only four of the other ten, which were treated with various weight-bearing methods, pro gressed to union.

Of the twelve patientswith delayedunion, one re

fused treatment, one was treated with a bone graft, and ten were treated initially by immobilization of the limb in a plaster cast and weight-bearing. Of these ten frac tures, seven healed in a mean of 15. 1 months and three eventually required grafting for non-union. Of the nine non-unions in the series, which were treated primarily with medullary curettage and bone-grafting, eight healed in a mean of three months. In all, twenty fractures were treated surgically with

an autogenous corticocancellous graft that was inlaid

after thorough curettage and drilling of the sclerotic bone that obliterated the intramedullary cavity. Of these twenty fractures, nineteen progressed to complete heal ing and one, to asymptomatic non-union. There were no other complications associated with the procedure. We concluded that the treatment of choice for acute fractures is immobilization of the limb in a toe-to-knee cast with non-weight-bearing. Fractures with delayed

unionmay eventuallyheal if they are treatedconser

* Sports Medicine Center. University of Pennsylvania. Weightman

Hall E7. 235 South 33rd Street. Philadelphia. Pennsylvania 19104. t Gannett Medical Clinic. 10 Central Avenue. Cornell University. Ithaca, New York 14833. Department of Radiology. The Hospital for Special Surgery. 535 East 70th Street. New York. N.Y. I0()21.

the proximal part of the diaphysis of the fifth metatarsal. All of the patients were evaluated clinically and roentgen ographically before and after treatment. Follow-up evalu ations were done in 1983 by another one of us (F. C. B.). and the forty-six fractures were followed for a mean of 40.2 months (range, six to 108 months). The fractures were classified and the results of treatment were evaluated ret rospectively after reviewing all clinic charts, hospital rec ords, and roentgenograms. All patients were interviewed at follow-up, sixteen by personal interview and examination and thirty by telephone, to obtain information regarding recurrence of symptoms, reinjury , refracture , and activity level . Treatment methods and results were then correlated with the fracture types. The mean age of all of the patients at the time of injury was 18.6 years, the youngest being fourteen and the oldest, twenty-nine years old. Thirty-nine of the forty-three patients were between the ages of sixteen and twenty-two years old.
209

VOL. 66-A, NO. 2. FEBRUARY 1984

210

J. S. TORG ET Al..

There were forty-one male and two female patients. All of the patients were engaged in some form of ath letic endeavor at the time of injury. Sixteen fractures were associated with playing basketball; fifteen, with football; six. with soccer: five, with baseball; and one fracture each, with gymnastics, lacrosse. field-hockey. and volleyball. The forty-six injuries were classified as acute fractures. delayed unions. or non-unions on the basis of the following clinical and roentgenographic criteria. The characteristic features of the acute fractures were: no history of previous fracture (although the patient may have had pain or discomfort), no intramedullary sclerosis (Fig. I-A), a fracture line with sharp margins and no widen ing or radiolucency. and minimum cortical hypertrophy or evidence of periosteal reaction to chronic stress. These roentgenographic features are not characteristlc of an acute fracture in the usual sense of the term. Presumably the acute fractures in our series were located at the site of a pre existing stress concentration or were in the lateral part of the cortex and became disabling when they extended across the entire diaphysis. Most important was the absence of intramedullary sclerosis (Fig. 1-A). The distinguishing features of the delayed unions were: a previous injury or fracture, or both; a fracture line that involved both cortices with associated periosteal new bone; a widened fracture line with adjacent radiolucency due to bone resorption: and evidence of intramedullary sclerosis (Fig. 2-B). The features of the non-unions were: a history of re petitive trauma and recurrent symptoms. a wide fracture line with periosteal new bone and radiolucency. and complete obliteration of the medullary canal at the fracture site by sclerotic bone (the hallmark of non-union) (Fig. 3-A). Twenty-six of the forty-six fractures were treated non

operatively with an orthosis with continued weight-bearing. immobilization in a plaster boot with continued weight bearing, or immobilization in a plaster boot with non weight-bearing. The other twenty fractures were treated with medullary curettage and an autogenous inlay bone graft for symptomatic delayed union or non-union. The criteria for a successful result after treatment were: ( 1) no symptoms. (2) roentgenographic evidence of solid union of the fracture, and (3) roentgenographic signs of recanalization of the niedullary canal with no medullary sclerosis. Twenty-one of the forty-six fractures in this study were described in a previous preliminary report@. They are in eluded in the present study because none had been cate gorized as to fracture type and none were tIeated by non weight-bearing immobilization. In the prev@uusreport It was concluded that these fractures alL slow to heal and often require bone-grafting. and that the initial treatment does not appear to influence the result@. Comparison of the results for the fractures that were treated initially with weight bearing and those for the fractures that were treated with non-weight-bearing is an important aspect of the present study. Also, in the previous report follow-up was terminated at the conclusion of both non-surgical and surgical treat ment, but Dameron and Kavanaugh et al. found that these patients have a tendency toward re-fractures. especially when they are treated surgically and followed for less than twenty-two months. Surgical Technique A consistent finding in the presence of non-union of this fracture was obliteration of the medullary canal by dense, sclerotic bone along the margins of the fracture. It is our opinion that the tendency of this fracture to progress

1:

ImTr

FIG. I-A

FIG. I-B

Fig. I .A: Oblique roentgenogram of the fifth metatarsal. demonstrating an acute fracture distal to the tuberosity. There is some cortical hypcrtrophy. an indicator ofchro)nic stress. but the fracture line is narrow. invo)lves both cortices. and. most important. is not associated ssith intramedullary sclerosis. Fig. 1-B: After treatment in a non-weight-bearing toe-to-knee cast for six weeks. there was complete healing. A roentgenogram niade nine months after the initial injury shos@smaintenance of fracture-healing.
THE JOURNAL OF BONE ANt) JOINT SURGERY

@
@

@I 1
FRAR.RES OF @I1EBASE OF THE FIFTH METATARSAL

211

@-:.

. .. ,@

.@ ..... .

.:::@@;.i..
:@

-@..

I'
@..

@4j
@
@ . -@ @-.

. ---

3%.@.

_.@k
. . .- $Vkr.@.i@

1 1@'3@V18
.@-, .-.

..

;.

Fiu. 2-A

Fi.

2-B

Fib. 2-C

Fig. 2-A: Oblique roentgcnogranl of the fifth metatarsal. demonstrating an acute fracture distal to the tuberosity. The patient was initially treated with a walking cast for six weeks. Fig. 2-B: 1-our months after injury the fracture line is seen to involve both cortices. and there is some angular deformity. a moderate degree of intramedullary sclerosis. and widening of the fracturc hue. Fig. 2-C': Two years hatci complete healing of the fracture has occurred with minimum deformity. and there is recanalization of the medullary canal.

to delayed union or non-union or to refracture after It has healed is due to the formation at the fracture site of this poorly organiied. sclerotic hone. vhich impairs healing and the strength of the union. lhc purpose of surgical tIcatnient is to: ( 1) re-establish the continuity of the niedullary canal by removing the sclerotic bone. and (2) facilitate healing of the fracture by insertIng an inlay boric graft. The base of the fifth metatarsal is approached through a curvilinear dorsolateral incision: the fracture site is cx posed subperiosteally; and a rectangulat section of bone measuring 0.7 by 2.0 centimeters. centered over the t'rac ture, is outlined by four drill-holes (Fig. 4-A) and removed with a sharp OsteOtOilie. The tiiedullary cavity is then cu retted or drilled. or both. until all of the sclerotic bone has been renioved and the continult\ of the iiicdullary canal has been re-established (Fig . 4-B . An autogenous corticocan cellous bone graft I1@easuring0.7 by 2.0 centimeters is then

removed from the anteromedial aspect of the distal end of the tibia through a second incision, being careful to contour the graft with a high-speed bun so that the cortical portion of the graft fits accurately into the rectangular cortical detect and does not protrude into the medullary canal and occlude it (Fig. 4-C). The periosteum, subcutaneous tissue, and skin
are closed sequentially in layers, and attention is turned to

the graft site in the tibia. To prevent the formation of a stress-raiser, the section of bone removed from the fracture site is placed in the tibial defect before the periosteurn. subcutaneous tissue, and skin are closed. A non-weight bearing plaster boot is applied. and immobilization is main tamed for six weeks. Results Of the forty-six injuries. twenty-five were classified as acute fractures: twelve, as delayed unions@and nine. as non

Fi(.

IA

FIG. 3-B

Fig. 3-A: Oblique roenigenogram demonstrating non-union. Note the s@idening of the fracture line. cortical hypertrophy. and dense intramedullary sclerosis completely obliterating the incdu1lai@ cavity. Fig. 3-B: Eight months after surgical treatment there is healing of the fracture and recanalization of the medullary canal.
VOL.. 66-A, NO. 2. FEBRUARY 1984

212

J. S. TORG ET At..

unions. Ten of the twenty-five acute fractures were treated with some form of immobilization and weight-bearing and the other fifteen. with immobilization and non-weight-bear ing. Of the ten acute fractures that were managed with weight-bearing (Figs. 2-A and 2-B), six were immobilized in a plaster boot for an average of 6.3 weeks (range, three to twelve weeks): three were supported by an orthosis, with activity being limited by pain; and one had no protection and activity was permitted as tolerated. Of these ten frac tures, asymptomatic delayed union developed in four, but they went on to clinical and roentgenographic healing after a mean of 11.5 months (range. seven to fifteen months) (Fig. 2-C). At follow-up. fifty-four to seventy-eight months (mean, seventy-one months) after injury. all four fractures were asymptomatic, and the patients were participating in full activities. Of the other six fractures, two had sympto matic delayed union and four, symptomatic non-union. All

operation. the patient was asymptomatic and engaging in full activities. Ofthe twelve fractures that were first seen with delayed union, ten were treated initially by immobilization and weight-bearing in a below-the-knee cast for four to nine weeks (mean, 6.0 weeks), one was treated surgically. and one patient refused treatment. Eight of the twelve fractures in this group eventually united at eight to twenty-six n@onths (mean, 14.8 months) after treatment. The other four, in eluding the one that was treated initially by operation and the three that failed to unite after treatment by weight-bear ing and immobilization. were treated surgically. All four of these fractures healed in twelve weeks. All twelve fractures
in the group were asymptomatic. and all patients were par

ticipating in full activities at follow-up at eleven to 108 months (mean, 44.3 months). All of the nine non-unions in the series were treated

Ft;.

4.A

FIG. 4-B

Fi.

4-C

Fig. 4-A: Suhperiosteal exposure of the base o)f the fifth metatarsal. distal to the tuberosity. through a dorsolateral curvilinear incision reveals the fracture line and associated cortical hvpertrophy. A rectangular piece of hone measuring approxitiiatelv 0.7 by 2.() centimeters. centered over the lateral aspect of the fracture. is outlined with four drill-holes.
Fig. 4-B: The piece of bone is excised with an osteotonie. and the sclerotic hone in the medullary canal is removed ss itli a curet or drill. or both. tO re-establish the continuity of the medullarv canal. Fig. 4-C: An autogenous cortical graft. obtained frotii the anteromedial aspect of the distal part of the tibia. is carefully contoured with a high-speed

burr and placed in the previously created defect. The periosteurn. subcutaneous
non-weight-bearing toe-to-knee cast is continued for six weeks.

tissues. and skin are then closed in layers. and imrnohiliiation

in a

six were treated with medullary curettage and bone-grafting, and all healed after twelve to fourteen weeks (mean. 12.6 weeks). At follow-up at six to fifty-eight months (mean, thirty-four months) after injury. all six were asymptomatic and the patients were engaging in full activities. Of the fifteen acute fractures that were initially treated by non-weight-bearing and immobilization in a plaster boot for an average of 6.5 weeks (range. six to nine weeks), fourteen healed after a mean of 7.4 weeks (range, six to twelve weeks), and the patients were asymptomatic and participating in full activities at the time of follow-up at twelve to seventy-six months (mean. 34.4 months) after injury (Figs. 1-A and 1-B). The remaining patient, whose acute fracture was treated in a non-weight-bearing below the-knee cast for six weeks, had a symptomatic non-union. The fracture healed twelve weeks after medullary curettage and bone-grafting. At follow-up, twenty-one months after

by us with medullary curettage and bone-grafting (Fig. 3-A). Eight healed in a mean of 12.2 weeks (range. ten to sixteen weeks) and one, in a patient who disregarded in structions not to bear weight postoperatively. did not heal. At the time of follow-up. eight to seventy-eight months after operation (mean. 32.7 months). one patient had an asymp tomatic non-union, one complained of vague discomfort in the foot despite roentgenographic evidence of healing of the fracture. and seven had asymptomatic healing of the frac ture. All patients were participating in full activities (Fig. 3-B). Of the twenty fractures in this series that were treated by bone-grafting for symptomatic delayed union or non union, nineteen (95 per cent) had healed, both clinically and roentgenographically, at twelve to sixteen weeks (mean, 12.3 weeks) after operation. In the remaining fracture there was a persistent non-union but it was asymptomatic at fol

FRACTURES OF THE BASE OF THE FIFTH METATARSAL

213

low-up. There were no distal tibial fractures through the donor site and no other complications. At the conclusion of treatment, forty-five fractures demonstrated solid union as well as resolution of medullary sclerosis roentgenographically. In no patient did a subse quent reinjury or re-fracture occur. At final follow-up all forty-three patients (forty-six fractures) were participating in full activities. With the exception of the one patient who had mild discomfort, all were asymptomatic. Discussion Fractures of the proximal end of the fifth metatarsal may involve the tuberosity or a 1.5-centimeter-long segment of bone distal to the tuberosity. It is important to appreciate this distinction because of the difference in the way that the two fractures respond to treatment. In a study by Dameron of 100 fractures involving the tuberosity that were treated with an elastic bandage. or with partial weight-bearing with crutches if the symptoms were severe, all but one had healed clinically at three weeks and roentgenographically at eight weeks. In contrast to these findings. Kavanaugh et al. , Dam eron. and Zelko et al. reported that fractures through the proximal part of the diaphysis are potentially disabling in juries in athletes. Our experience with forty-six such frac tures is in agreement with this observation. Jones. in 1902, was the first to describe diaphyseal fractures of the fifth metatarsal, when he reported on four such injuries. all of which apparently healed with conser vative management. In 1927, Carp recognized that these fractures tend to heal poorly; he found that of twenty-one fractures union was delayed in five. In 1960, Stewart noted that bone-grafting was needed to secure union of some of these fractures, and also that there may be considerable
variation injury. twenty in the duration Dameron diaphyseal reported fractures and that severity five (25 of symptoms per cent) required after of the

in his series

bone-graft

ing for symptomatic non-union. Kavanaugh et al. noted that in their series union was delayed in twelve (66.7 per cent) of eighteen fractures that were treated conservatively. In their total group of twenty-three fractures, thirteen (56.5 per cent) were eventually treated surgically using an intra medullary screw for fixation. In our series symptomatic delayed union or non-union developed in twenty (43.5 per cent) of forty-six fractures and was treated surgically. Of Dameron's twenty patients. nine were treated with a plaster cast and eleven, with elastic bandages and crutches. Fifteen of the fractures had conservative treatment only, with twelve healing at between two and twelve months after treatment and the other three healing after fifteen, twenty, and twenty-one months. The remaining five patients in the series were treated with a sliding bone graft for symptomatic non-union at four, seven, eight, fourteen, and fifteen months. Based on these findings, Dameron concluded that initial treatment did not appear to influence the result. Sim ilarly, Kavanaugh et al. found that several of their patients who were treated with non-weight-bearing and a plaster cast for ten to twelve weeks showed no evidence of healing, and
VOL. 66-A, NO. 2. FEBRUARY i984

they concluded that plaster immobilization and non-weight bearing were probably unnecessary. Zelko et a). , in a pre liminary report on the first twenty-one patients in our series (none of whom had been treated with non-weight-bearing), concluded that the clinical course did not appear to be influ enced by the usual conservative treatment. Subsequent to the report of Zelko et al. our series of fractures were classified as already described, and fifteen of the acute ones were treated with non-weight-bearing in a plaster cast. Fourteen of the fifteen healed in a mean of 6.5 weeks. This finding is in sharp contrast to that in the initial ten acute fractures, which were all treated with a weight-bearing cast. Of these ten fractures. four healed in an average of 11.5 months, while asymptomatic delayed union or non-union developed in the other six and bone grafting was necessary. The failure of the fractures to unite in the patients of Kavanaugh et al. , despite treatment with a non-weight-bear ing cast, may be explained by the fact that they were not acute fractures but were reinjuries in patients with delayed union that required either prolonged immobilization or bone grafting. On the basis of our experience with acute fractures of the base of the fifth metatarsal distal to the tuberosity. we believe that immobilization in a non-weight-bearing plaster boot is the treatment of choice. When there is a delayed union it is apparent that, given enough time (mean, 14.8 months in our series), union will occur in virtually all pa tients if vigorous activity is avoided. However, for athletes who desire to return to competition non-operative treatment usually will not be acceptable. As described in the literature, delayed union and non union are treated by two basic surgical techniques: ( 1) in tramedullary fixation using a screw, and (2) a sliding or inlay bone graft. Common to both of these methods is re moval of the sclerotic bone plug from the medullary canal. It is our belief that the poorly organized, sclerotic intra medullary bone that forms when union is slow contributes to the frequent delay in healing, re-fracture, and non-union after this injury. Complete union is achieved only when the medullary canal is recanalized. We believe that intramed ullary curettage hastens this process in vigorous, physically active individuals. Intramedullary screw fixation is associated with several problems. To approach the lesion from the proximal end of the tuberosity and place a Leinbach or an AO malleolar
screw, as noted by Kavanaugh et al. , is technically difficult.

Also, as reported by these authors, this technique is asso ciated with a 45 per cent rate of perioperative complications. In their thirteen patients, three intramedullary screws frac tured at operation, two missed the medullary canal, and one had to be removed because ofpain. Intramedullary curettage and inlay bone-grafting, on the other hand, are technically simple, are not associated with perioperative complications, and have resulted in a 95 per cent rate of union. We believe that it is the procedure of choice for delayed union or non union of these fractures.

214 Conclusions

J. S. TORG ET AL. Of intramedullary sclerosis partially obliterating the med

On the basis of our experience. it appears that the selection of the best form of treatment for these fractures should be based on the presence or absence of medullary sclerosis adjacent to the fracture site. In our series, fourteen (93 per cent) of the fifteen fractures with no associated intramedullary sclerosis were successfully treated with non
weight-bearing and immobilization in a plaster boot for six

to eight weeks. We therefore believe that this is the treatment of choice for fractures involving the base of the fifth metatar sal distal to the tuberosity when roentgenograms demon strafe no intramedullary sclerosis. On the other hand, if roentgenograms show evidence

ullary cavity and if the patient engages in non-jumping ac tivities, such as football, baseball, or soccer, it may take a long time for the fracture to heal or for an asymptomatic delayed union to develop. Thus, in serious amateur or professional athletes, it might be preferable to expedite heal ing by surgical means. If roentgenograms demonstrate non-union with dense sclerotic bone adjacent to the fracture line and a completely obliterated medullary cavity, medullary curettage and au togenous inlay bone-grafting are recommended for all ath letes and should be considered in non-athletes who are symptomatic during their usual activities.

References
1. CARP, Louis: Fracture of the Fifth Metatarsal Bone. With Special Reference to Delayed Union. Ann. Surg. . 86: 308-320. 1927. 2. DAMERON.T. B. . JR. : Fractures and Anatomical Variations of the Proximal Portion of the Fifth Metatarsal. J. Bone and Joint Surg. . 57-A: 788792, Sept. 1975.

3. JONES. ROBERT: Fracture of the Base of the Fifth Metatarsal Bone by Indirect Violence. Ann. Surg. . 35: 697-700, 1902. 4. KAVANAUGH.J. H.; BROWER, 1. D.; and MANN. R. V.: The Jones Fracture Revisited. J. Bone and Joint Surg. . 60-A: 776-782. Sept. 1978. 5. STEWART. I. M.: Jones's Fracture: Fracture of the Base of the Fifth Metatarsal. Clin. Orthop. . 16: 190-198, 1960.
6. ZELKO, R. R. : TORG. J. S. ; and RACHUN. ALEXIS: Proximal Diaphyseal Fractures of the Fifth Metatarsal Treatment of the Fractures and Their

Complications

in Athletes. Am. J. Sports Med. , 7: 95-101 . 1979.

THE JOURNAL OF BONE AND JOINT SURGERY