Downloaded from www.ajronline.org by 96.246.45.159 on 10/16/13 from IP address 96.246.45.159. Copyright ARRS. For personal use only; all rights reserved

Morton’s Neuroma: Is It Always Symptomatic? Jenny Bencardino 1 Zehava S. Rosenberg 2 Javier Beltran 3 Xiang Liu 4 Emmanuelle Marty-Delfaut 2

OBJECTIVE. We determined the prevalence of clinically silent Morton’s neuroma and searched for distinguishing MR imaging features of Morton’s neuroma in patients with clinical complaints related to this entity and in patients with clinically silent lesions. MATERIALS AND METHODS. One radiologist who was unaware of clinical findings retrospectively reviewed 85 consecutive foot MR examinations. MR imaging criteria for Morton’s neuroma included a low- to intermediate-signal-intensity soft-tissue mass in the intermetatarsal space. The size, location, and signal intensity of each neuroma and the presence of intermetatarsal bursae were recorded. The patients were subdivided into symptomatic or asymptomatic groups on the basis of the patients’ answers on a questionnaire documenting the locations and characteristics of symptoms and discussions with each referring physician about clinical findings. Surgical confirmation was available in eight of 25 symptomatic patients. RESULTS. The prevalence of Morton’s neuroma in patients with no clinical evidence of this condition was 33% (19/57). Twenty-five patients had symptomatic Morton’s neuroma, 19 had Morton’s neuroma based on MR imaging findings with no clinical manifestations, and 41 did not have Morton’s neuroma. Slightly larger lesions were observed in the symptomatic group, although significant overlap was noted between the two groups. The mean transverse diameter of symptomatic neuromas was 5.3 mm (standard deviation, 2.14) compared with 4.1 mm (standard deviation, 1.75) for asymptomatic neuromas; this difference was marginally significant (p = 0.05). CONCLUSION. The MR imaging diagnosis of Morton’s neuroma does not imply symptomatology. Careful correlation between clinical and MR imaging findings is mandatory before Morton’s neuroma is considered clinically relevant.

M

Received August 5, 1999; accepted after revision February 16, 2000. 1 Department of Radiology, Massachusetts General Hospital, 15 Parkman St., Ste. 515, Boston, MA 02114. Address correspondence to J. Bencardino. 2

Department of Radiology, Hospital for Joint Diseases-OI, 305 E. 17th St., New York, NY 10003. 3

Department of Radiology, Maimonides Medical Center, 4802 10th Ave., Brooklyn, NY 11219.

4 Department of Radiology, Long Island Jewish Medical Center, 270-05 76th Ave., New Hyde Park, NY 11040.

AJR 2000;175:649–653 0361–803X/00/1753–649 © American Roentgen Ray Society

AJR:175, September 2000

orton’s neuroma is a common cause of forefoot pain. The exact cause of this nonneoplastic condition is not well known, although several theories have been proposed, including ischemia and mechanical compression of the plantar nerve against the transverse intermetatarsal ligament [1]. Histologic findings of Morton’s neuroma include neural degeneration, epineural and endoneural vascular hyalinization, and perineural fibrosis surrounding an intermetatarsal nerve frequently located in the third intermetatarsal space, in which the medial and lateral plantar nerves approximate [2] (Fig. 1). The condition is most commonly seen in middle-aged patients. Sonography, CT, and MR imaging have been used in the examination of patients with clinical findings suggestive of Morton’s neuroma [3, 4]. Nonsurgical treatment consists of modification of patient’s foot-

wear, taping, padding, orthoses, and steroid injections. Surgery is indicated when conservative measures fail to relieve symptoms [5]. MR imaging has proven useful in the diagnosis of Morton’s neuroma when the location of pain is atypical or when imaging is required before surgery [6–8]. The occurrence of findings suggestive of Morton’s neuroma on MR imaging in asymptomatic volunteers has recently been described [9]. We assessed, on the basis of MR imaging of the foot, the prevalence of asymptomatic Morton’s neuroma; additionally, we determined whether MR imaging reveals differences between symptomatic and clinically silent lesions. Materials and Methods Ninety-five consecutive MR examinations of the foot performed during a 24-month period were retrospectively reviewed. Ten MR imaging examina-

649

Downloaded from www.ajronline.org by 96.246.45.159 on 10/16/13 from IP address 96.246.45.159. Copyright ARRS. For personal use only; all rights reserved

Bencardino et al.

Fig. 1.—Drawing shows intermetatarsal nerve. Note third intermetatarsal nerve passing underneath intermetatarsal ligament.

tions were excluded because of poor image quality (motion artifacts or inadequate coverage) or the presence of an abnormality in the area of the metatarsal heads (osteomyelitis, fracture, and pigmented villonodular synovitis). Eighty-five patients were included in our study (21 males and 64 females; age range, 16–86 years; mean age, 48 years). The patients were subdivided into symptomatic or asymptomatic groups on the basis of the patients’ answers on a questionnaire documenting the location and characteristics of symptoms and discussions with each referring physician about clinical findings. Clinical symptoms compatible with Morton’s neuroma included pain or paresthesias in the intermetatarsal region worsened by weight-bearing or the use of high-heeled shoes, often relieved by removing shoes and gentle massage of the forefoot. On physical examination, a positive web space compression test or a painful palpable click (Mulder’s sign) was considered suggestive of Morton’s neuroma. Twenty-eight patients had clinical signs and symptoms suggestive of Morton’s neuroma. Fiftyseven patients underwent MR imaging of the foot for indications other than Morton’s neuroma, including hind-, midfoot, or metatarsal fracture (n = 12); palpable mass in regions other than the intermetatarsal area (n = 12); follow-up status after surgery on the hind- or midfoot (n = 7); painful great toe (n = 7); tendinous tear (n = 7); painful hind- or midfoot (n = 6); diffuse foot swelling (n = 2); osteomyelitis (n = 2); and sinus tarsi syndrome (n = 2). Surgical confirmation was available in eight of 25 patients with clinical symptoms and MR findings suggestive of Morton’s neuroma. MR images were obtained on a 1.5-T scanner (Signa; General Electric Medical Systems, Milwaukee, WI) using a transmit–receive extremity

650

coil. All patients included in our study underwent T1-weighted (TR range/TE range, 500–900/15–2) and fast spin-echo T2-weighted (TR range/TE range, 3000–3800/95–100) oblique coronal (perpendicular to the metatarsal bones) MR imaging and T1-weighted (TR/TE, 600/15) oblique axial (parallel to the metatarsal bones) MR imaging. The field of view was 12–14 cm, and the slice thickness was 3–4 mm with an interslice gap of 1 mm. The image matrix was 512 × 256 for T1-weighted MR images and 256 × 256 for T2-weighted and T1weighted fat-suppressed MR images. In eight patients, axial gradient-echo MR images (TR range/TE range, 400–500/15–20; flip angle, 20°) were also obtained. In another eight patients, T1-weighted fat-suppressed spin-echo axial MR images (TR/TE, 700/15) were obtained after the injection of 0.1 mmol of gadopentetate per kg body weight. Fourteen of these patients had clinical findings suggestive of Morton’s neuroma, and the aforementioned sequences were obtained to further characterize MR imaging findings already noted on standard protocol images. The retrospective review of 85 consecutive foot MR examinations for the presence or absence of Morton’s neuroma was performed by an experienced musculoskeletal radiologist who was unaware of the clinical findings. The diagnosis of Morton’s neuroma was determined on the basis of previously established MR imaging criteria [6–8]: neuromas were diagnosed when the radiologist noted a soft-tissue mass within the intermetatarsal space that had low to intermediate signal intensity on T1-weighted MR images and a more variable but usually low to intermediate signal intensity on T2-weighted MR images. The transverse medial–lateral diameter of lesions was measured on oblique coronal T1weighted MR images (perpendicular to the metatarsal bones) using calipers calibrated to scale. The minimum mass size was 2 mm. The difference in mean transverse size of silent versus symptomatic Morton’s neuromas was assessed using an unpaired two-tailed Student’s t test. The differences in plantar extension, signal intensity, and location between the silent and symptomatic Morton’s neuromas were measured using chi-square analysis.

TABLE 1

The location and signal intensity characteristics on T1- and T2-weighted MR images were recorded for each neuroma. Signal behavior on gradient-echo T2*-weighted images and fat-suppressed gadolinium-enhanced T1-weighted MR images was also annotated in 16 patients. Diagnostic criteria for distended intermetatarsal bursa included well-defined areas greater than 1 mm in transverse diameter with low signal intensity on T1-weighted MR images and bright signal intensity on T2-weighted MR images obtained in the intermetatarsal area.

Results

On the basis of their clinical presentation, 44 patients with positive findings on MR imaging for Morton’s neuroma were divided into two groups: clinically symptomatic Morton’s neuroma and clinically silent Morton’s neuroma. In the following text, we describe the MR imaging findings of each group. Clinically Symptomatic Morton’s Neuroma

Twenty-eight of 85 patients presented with symptoms and signs of Morton’s neuroma on clinical examination. Three of these patients did not reveal features of Morton’s neuroma on MR imaging. MR imaging revealed a bone contusion of the first metatarsal head in one of these patients. A second patient had findings suggestive of inflammatory synovitis of the metatarsal phalangeal joint. The last patient had third-space intermetatarsalgia based on clinical findings; the patient’s MR imaging was unremarkable, and the patient was lost to follow-up. Thirty-three Morton’s neuromas were diagnosed in 25 symptomatic individuals (four men and 21 women; mean age, 49 years) (Table 1). Nineteen of the lesions were encountered in the third web space, 12 were in the second web space, and two were in the first web space. Seven lesions had plantar exten-

Comparative Analysis of MR Imaging Findings of 55 Morton’s Neuromas Finding

Plantar extension Low signal intensity on T1-weighted MR images High signal intensity on T2-weighted MR images Location First web space Second web space Third web space Fourth web space

Asymptomatic Symptomatic Morton’s Neuroma Morton’s Neuroma

p

2/22 (9) 22/22 (100) 5/22 (22)

7/33 (21) 33/33 (100) 8/33 (24)

0.05 — —

3/22 (14) 9/22 (41) 10/22 (45) 1/19 (5)

2/33 (6) 12/33 (36) 19/33 (58) 2/25 (8)

— — — —

Note.—Numbers in parenthesis are percentages. Dash (—) indicates not significant.

AJR:175, September 2000

Downloaded from www.ajronline.org by 96.246.45.159 on 10/16/13 from IP address 96.246.45.159. Copyright ARRS. For personal use only; all rights reserved

Morton’s Neuroma

A

sion beyond the inferior border of the metatarsal head. Eight symptomatic patients (32%) had coexistent Morton’s neuromas in adjacent intermetatarsal spaces. In one patient, the lesions were in the first and second web spaces. The other seven patients had coexistent second and third interdigital neuromas. Fourteen of these 16 concurrent neuromas showed identical signal intensity characteristics. In four of these concurrent neuromas, one of the lesions was 2 mm or larger than the second one. The larger lesions were all located in the second intermetatarsal space. All the lesions appeared hypointense to adjacent fat on T1-weighted MR images. Eight (24%) of 33 symptomatic Morton’s neuromas revealed intermediate to mild hyperintensity to fat on T2-weighted MR images. In 10 of these lesions, additional gradient-echo MR images were obtained that revealed marked hyperintensity of the neuromas (Fig. 2). Intense homogeneous enhancement was noted in six of seven patients who received IV injection of gadolinium. The size of the lesion ranged from 3 to 10 mm (mean, 5.3 mm; standard deviation, 2.14). Two fluid-filled intermetatarsal bursae were identified. One bursa was in the same location as a neuroma; the other one was in a different web space (Fig. 2). Asymptomatic Morton’s Neuroma

B

C Fig. 2.—35-year-old symptomatic woman with surgically proven Morton’s neuroma. A, Oblique coronal T1-weighted MR image (TR/TE, 850/15) shows well-demarcated 10-mm hypointense soft-tissue mass (arrows) in second intermetatarsal space, extending plantarly. B, Oblique coronal T2-weighted fast spin-echo MR image (3500/100) shows areas of slightly increased signal intensity in neuroma (arrows). Note small fluid-filled intermetatarsal bursa located in first intermetatarsal space (asterisk). C, Oblique coronal gradient-echo T2*-weighted MR image (500/15; flip angle, 20°) shows homogeneous increase in signal intensity of mass (arrows) in second intermetatarsal space.

AJR:175, September 2000

Nineteen of 57 patients with no clinical complaints of Morton’s neuroma (six men and 13 women; mean age, 47) had findings suggestive of Morton’s neuroma based on MR imaging criteria. Twenty-two neuromas were identified in these 19 patients (Table 1). All the lesions appeared hypointense on T1weighted MR images. Seventeen neuromas revealed low signal intensity, and five revealed intermediate to hyperintense signal intensity on T2-weighted MR images. One of the patients had IV injection of contrast material without evidence of enhancement. Ten of the lesions were identified in the third intermetatarsal space, nine in the second, and three in the first. The neuromas were confined to the intermetatarsal space in 20 patients. Two lesions extended below the interior border of the metatarsal heads in the second and third web spaces (Fig 3). Three patients (16%) had concomitant neuromas in two different web spaces. Four of these lesions revealed identical signal behavior on T1- and T2-weighted MR imaging. One patient had signal discrepancy in one pair of neuromas; one of the lesions was brighter on T2-weighted MR images. Size differences

651

Downloaded from www.ajronline.org by 96.246.45.159 on 10/16/13 from IP address 96.246.45.159. Copyright ARRS. For personal use only; all rights reserved

Bencardino et al.

A

B Fig. 3.—41-year-old woman without clinical symptoms of Morton’s neuroma. A and B, Oblique coronal T1-weighted MR image (TR/TE, 900/20) (A) and oblique coronal fast spin-echo T2weighted MR image (3500/100) (B) obtained through intermetatarsal region depict 7-mm low-signal-intensity soft-tissue mass located in third web space (arrows).

greater than 2 mm were not observed in the silent concomitant neuromas. The lesions ranged in size from 2 to 9 mm (mean, 4.1 mm; standard deviation, 1.75). One distended bursa was identified in the third intermetatarsal space. This fluid collection was not associated with a neuroma in the same location. Statistical Results

The incidental MR imaging diagnosis of Morton’s neuroma was made in 33% of patients (19/57) with no clinical complaints of this condition. An unpaired two-tailed Student’s t test revealed marginal statistically significant differences in the size of Morton’s neuromas between the clinically symptomatic and clinically silent groups of patients (p = 0.05). No statistically significant differences

652

were found between the two groups in regard to plantar extension (χ 2 = 1.417), low signal intensity on T1-weighted MR images (equal percentages), high signal intensity on T2weighted MR images (χ 2 = 0.017), first web space location (χ 2 = 0.916), second web space location (χ 2 = 0.1154), or third web space location (χ 2 = 0.879). Fluid-filled intermetatarsal bursae were found in both groups without a statistically significant difference (χ 2 = 0.127).

Discussion

Intermetatarsal neuromas were first reported by Durlacher in 1845 [10]; however, the term Morton’s neuroma was popularized after the description of this condition by Thomas G. Morton in 1876 [11]. Initially,

the lesion was thought to exclusively affect the third intermetatarsal space. A thicker third intermetatarsal nerve formed by the fusion of the medial and lateral plantar nerves is more easily entrapped by the overlying intermetatarsal ligament [5]. However, later studies revealed that other intermetatarsal spaces can also be affected [6–8]. In our series, the third and second spaces were the most commonly involved sites. Although the exact cause is unknown, Morton’s neuroma most likely represents an entrapment neuropathy causing perineural fibrosis, nerve degeneration, leukocyte infiltration, and epineural and endoneural vascular hyalinization that results in a significantly thicker intermetatarsal nerve [2, 12, 13]. Such thickening can create further trauma that results in more thickening and subsequently more entrapment. Excessive weight bearing on the forefoot related to the use of high-heeled shoes has also been implicated and may explain the higher prevalence of Morton’s neuroma in middle-aged women [1]. Despite its benign nature, Morton’s neuroma can induce enough discomfort to make walking difficult. Patients classically experience sharp pain, a burning sensation, and paresthesias during weight bearing in the region of the intermetatarsal spaces. The pain is relieved by rest and shoe removal. On physical examination, a mass can be palpated in one third of patients. This finding is often accompanied by a characteristic click or Mulder’s sign [5]. In some patients, the diagnosis may be equivocal, and other causes of intermetatarsalgia may be clinically entertained. The differential diagnosis includes intermetatarsal bursitis, true neuroma, inflammatory arthritis, pigmented villonodular synovitis, osteomyelitis, foreign body granuloma, stress fracture, Freiberg’s infraction, and metatarsophalangeal joint dislocation. In these instances, MR imaging has proven valuable in establishing the underlying cause of intermetatarsal pain [6–8]. Additionally, if surgery is considered, MR imaging can confirm the diagnosis and reveal the number and exact location of neuromas. Because of its predominantly fibrous composition, Morton’s neuroma appears on T1and T2-weighted spin-echo MR images with low to intermediate signal intensity [6, 8]. Other soft-tissue masses, including fibromatosis, nodular fasciitis, and pigmented villonodular synovitis, may exhibit a similar signal pattern related to low cellularity, high collagen content, or hemosiderin deposits. However, hypointensity on T1- and T2-weighted MR images, combined with the typical location

AJR:175, September 2000

Downloaded from www.ajronline.org by 96.246.45.159 on 10/16/13 from IP address 96.246.45.159. Copyright ARRS. For personal use only; all rights reserved

Morton’s Neuroma and a distinct clinical history, is characteristic of Morton’s neuroma. Some investigators have advocated the use of short tau inversion recovery (STIR) and gadolinium to increase the lesion’s conspicuity on MR imaging [7, 14]. Zanetti et al. [8] compared standardized contrast-to-noise ratios of T1-weighted spin-echo, T2-weighted turbo spin-echo, STIR, and T1-weighted fat-suppressed spin-echo MR images after gadolinium injection. T1-weighted MR images had the highest contrast-to-noise ratio of all four sequences, and T2-weighted MR images were useful in the evaluation of other diagnostic possibilities that could reveal high signal intensity on T2-weighted MR images, such as true neuromas [15], intermetatarsal bursitis [9], ganglion cysts, and synovial cysts [16]. STIR and enhanced T1-weighted fat-suppressed MR images did not provide significant additional information to justify their routine use in the evaluation of Morton’s neuromas [8]. Other investigators [17] have also questioned the usefulness of enhanced T1-weighted fat-suppressed sequences in the assessment of this condition. In our series, all silent and symptomatic lesions were readily identified on T1-weighted MR images as low-signal-intensity masses when compared with adjacent brighter fat tissue (Figs. 2 and 3). Most neuromas appeared with low signal intensity on T2-weighted MR images. However, eight (24%) of 33 symptomatic and five (22%) of 22 silent lesions showed increased signal intensity on T2-weighted MR images (Fig. 2C). This T2 prolongation may be explained by acute ischemic changes and leukocyte infiltration of the intermetatarsal nerve, which causes endoneural and epineural edema [13]. Although this hypothesis may be a reasonable pathophysiologic explanation for this phenomenon, it did not correlate with the presence or absence of clinical symptoms in our study. Marked hyperintensity was seen in 10 of 10 lesions on gradient-echo T2*-weighted MR images and in six of seven lesions in patients who received an IV injection of gadolinium. However, we believe that the diagnosis of Morton’s neuroma in symptomatic patients could be made on the basis of the information obtained from standard T1- and T2-weighted fast spin-echo images, provided the images show the typical signal behavior and location of Morton’s neuromas.

AJR:175, September 2000

The only marginally significant difference between silent and symptomatic Morton’s neuromas was a slightly greater mean transverse diameter of the lesion in the symptomatic group of patients. In the clinically silent group, only four of 22 lesions measured more than 5 mm. The mean transverse diameter of the silent lesions was 4.1 mm compared with 5.3 mm for symptomatic neuromas. This finding is consistent with the proposed cause of entrapment neuropathy: the thicker the lesion, the greater the risk of being entrapped by the overlying intermetatarsal ligament. Similar results were previously reported by Zanetti et al. [9]. These investigators established a threshold of 5 mm (in transverse diameter) beyond which the lesion is most likely symptomatic. Plantar extension of the lesion further below the level of the metatarsal head was observed in seven of 33 symptomatic masses and in two of 22 silent masses (Figs. 2 and 3). We noted no statistically significant difference between the two groups for the presence of plantar extension beyond the inferior margin of the metatarsal heads. Distended intermetatarsal bursae were found in patients with symptomatic and clinically silent Morton’s neuromas. Distended intermetatarsal bursae were encountered in only 7% of our patients, in contrast to previous reports of a higher prevalence in asymptomatic volunteers [9]. Many of the lesions included in our series were not surgically confirmed; this may be considered a limitation of our study. However, the MR imaging criteria for Morton’s neuroma are well established; therefore, we do not believe this factor represents a major limitation. Two other potential problems include the retrospective study design and the inaccurate size assessment of some neuromas (occasionally, some lesions had irregular and poorly defined margins). In summary, on the basis of MR imaging, we diagnosed Morton’s neuroma in 33% (19/57) of patients with no clinical evidence of this condition. Slightly larger lesions were observed in the symptomatic group of patients; however, significant overlap was noted between the two groups. On the basis of these results, we conclude that MR imaging findings suggestive of Morton’s neuroma do not imply symptomatology and that careful correlation between clinical and MR imaging findings is

mandatory before a Morton’s neuroma is considered clinically relevant. References 1. Wu KK. Morton’s interdigital neuroma: a clinical review of its etiology, treatment and results. J Foot Ankle Surg 1996;35:187–188 2. Graham CE, Graham DM. Morton’s neuroma: a microscopic evaluation. Foot Ankle 1984;5:150–153 3. Shapiro PP, Shapiro SL. Sonographic evaluation of interdigital neuromas. Foot Ankle Int 1995;16:604–606 4. Turan I, Lindgren U, Sahlstedt T. Computed tomography for diagnosis of Morton’s neuroma. J Foot Surg 1991;30:244–245 5. Medicino SS, Rockett MS. Morton’s neuroma: update on diagnosis and imaging. Clin Podiatr Med Surg 1997;14:303–311 6. Erickson SJ, Canale PB, Carrera GF, et al. Interdigital (Morton) neuroma: high resolution MR imaging with a solenoid coil. Radiology 1991; 181:833–836 7. Terk MR, Kwong PK, Suthar M, et al. Morton neuroma: evaluation with MR imaging performed with contrast enhancement and fat suppression. Radiology 1993;189:239–241 8. Zanetti M, Ledermann T, Zollinger H, et al. Efficacy of MR imaging in patients suspected of having Morton’s neuroma. AJR 1997;168:529–532 9. Zanetti M, Strehle JK, Zollinger H, Hodler J. Morton neuroma and fluid in the intermetatarsal bursae on MR images of 70 asymptomatic volunteers. Radiology 1997;203:516–520 10. Durlacher L. A treatise on corns, bunions, the disease of nails, and the general management of the feet. Philadelphia: Lea and Blanchard, 1845:52 11. Morton TG. A peculiar and painful affection of the fourth metatarsophalangeal articulation. Am J Med Sci 1876;71:37–45 12. Bourke G, Owen J, Machet D. Histological comparison of the third interdigital nerve in patients with Morton’s metatarsalgia and control patients. Aust N Z J Surg 1994;64:421–424 13. Shereff MJ, Grande DA. Electron microscopic analysis of the interdigital neuroma. Clin Orthop 1991;271:296–299 14. Unger HR Jr, Mattoso PQ, Drusen MJ, Neumann CH. Gadopentetate-enhanced magnetic resonance imaging with fat saturation in the evaluation of Morton’s neuroma. J Foot Surg 1992;31:244–246 15. Hoskins CL, Sartoris DJ, Resnick D. Magnetic resonance imaging of foot neuroma. J Foot Surg 1992;31:10–16 16. May DA, Good RB, Smith DK, Parsons TW. MR imaging of musculoskeletal tumor mimickers with intravenous gadolinium: experience with 242 patients. Skeletal Radiol 1997;26:2–15 17. Williams JW, Meaney J, Whitehouse GH, Klenerman L, Hussein Z. MRI in the investigation of Morton’s neuroma: which sequences? Clin Radiol 1997;52:46–49

653