| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1 Radiologisches Institut, Universitätsspital Basel, Petersgraben 4, 4031
Basel, Switzerland.
2 Department of Radiology, Thomas Jefferson University Hospital, 111 S. 11th
St., No. 3390, Gibbon, Philadelphia, PA 19107.
3 Department of Othopedic Surgery, Rothman Institute, Thomas Jefferson
University Hospital, 925 Chestnut St., Philadelphia, PA 19107.
Received January 28, 2002;
accepted after revision April 1, 2002.
Address correspondence to H. P. Ledermann.
Abstract
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionAPPENDIX 1. MR Imaging...References |
|---|
MATERIALS AND METHODS. Contrast-enhanced MR imaging examinations of 159 infected feet performed at 1.5 T were reviewed by two musculoskeletal radiologists for the presence and location of tendon infection (peritendinous enhancement contiguous to an adjacent ulcer or cellulitis) and for the spread of infection along tendons, which was defined as peritendinous contrast enhancement extending more than 2 cm beyond surrounding cellulitis. The study group was composed of 156 consecutive patients, 82.7% of whom had diabetes; all patients underwent subsequent surgical treatment. Results of MR evaluations were compared with the patients' charts and surgical reports.
RESULTS. Of the 129 MR examinations showing an infection in the forefoot, MR evidence of tendon involvement in the infection was observed in 56 MR examinations (43%). The sum of involved tendons per ray was as follows: first ray (flexor tendon, n = 19; extensor tendon, n = 13), second (flexor tendon, n = 12; extensor tendon, n = 7), third (flexor tendon, n = 5; extensor tendon, n = 4), fourth (flexor tendon, n = 5; extensor tendon, n = 1), and fifth (flexor tendon, n = 11; extensor tendon, n = 6). Of the 32 MR examinations showing infection in the hindfoot, 14 examinations (44%) showed evidence of tendon involvement, most frequently of the distal Achilles tendon (n = 7). MR evidence of the spread of infection along a tendon was seen in 12 examinations, always with proximal spread of infection; and infection led to the development of an abscess in the central plantar compartment in three patients. Intraoperative evidence of a tendon infection was documented in 11 patients. The surgical procedure was altered because of the tendon infection in six patients.
CONCLUSION. MR evidence of tendon infection is present in approximately half the patients who require surgery for pedal infection. Evidence of spread of the infection along tendons is seen infrequently on MR imaging. Detection of a tendon infection could influence surgical therapy.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionAPPENDIX 1. MR Imaging...References |
|---|
The goal of this study was to evaluate the frequency and distribution of tendon involvement in pedal infection and the frequency of spread of infection along tendons in patients with advanced pedal infection who underwent surgical intervention after MR evaluation.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionAPPENDIX 1. MR Imaging...References |
|---|
MR imaging was performed with a 1.5-T super-conducting magnet (Signa; General Electric Medical Systems, Milwaukee, WI). An extremity coil was used for 151 of the 159 feet (field of view, 14-20 cm); a head coil was used for eight patients who underwent imaging of both feet (field of view, 16-20 cm). In these eight patients, only the eight feet that underwent subsequent surgery were included in the study. Images of all feet were obtained in at least two orthogonal planes.
T1-weighted spin-echo MR images were obtained with 1-2 signals averaged, a TR range/TE range of 400-750/10-20, and a matrix of 256 x 192 or 256 x 256. T2-weighted images were obtained using a fast spin-echo technique with 2 signals averaged, an echo-train length of 8, a TR range/TEeff range of 2000-7800/75-108, and a matrix of 256 x 128 or 256 x 192. Contrast-enhanced fast-suppressed T1-weighted images of 31 feet were obtained using a spin-echo sequence without obtaining unenhanced fat-suppressed T1-weighted images. For 128 feet, fast multiplanar spoiled gradient-recalled imaging with a TR/TE of 250/2.1, a flip angle of 90°, and a matrix of 256 x 128 or 256 x 192 was performed using the same parameters before and after contrast administration. An IV contrast agent (Magnevist [gadopentetate dimeglumine]; Berlex Laboratories, Wayne, NJ) was administered at a dose of 0.1 mmol/kg of body weight. For all T2-weighted and gadolinium-enhanced T1-weighted sequences, fat suppression was accomplished using selective presaturation of lipid resonant frequency.
Fast spin-echo short tau inversion recovery images were obtained with an echo-train length of 8, a TR range/TEeff range of 3000-6000/20-78, an inversion time of 150-160 msec, and a matrix of 256 x 128 or 256 x 192. Fast spin-echo short tau inversion recovery images were available for 136 feet.
Our MR imaging criteria for a tendon infection, cellulitis [5], skin ulcer [2], peritendinous spread of infection, abscess [7, 8], and osteomyelitis [1, 9, 10] are summarized in Appendix 1. Surrounding peritendinous fluid signal was not a prerequisite to diagnose tendon infection because not all pedal tendons are covered by a tendon sheath [11] and because inflammatory destruction of a tendon sheath may result in the leakage of synovial fluid.
Two musculoskeletal radiologists reviewed in consensus all MR studies for the presence of tendon infection. Both reviewers knew that all the patients had a foot infection, but they were unaware of all other clinical data. The riviewers noted which tendon or tendons were involved and recorded the location and extension of infection of each involved tendon on a schematic anatomic drawing of the foot in the anteroposterior and lateral projections. The location of the tendon infection was defined using the following anatomic regions: toes, metatarsophalangeal joint, proximal forefoot (from the metatarsophalangeal joint to the Lisfranc's joint), and hindfoot. The following information was also noted: whether the diameter of the involved tendon (i.e., tendon proper without tendon sheath) was thickened relative to its size proximal and distal to the infection, whether the tendon had a hyperintense signal on T2-weighted images, whether a hyperintense rim could be seen around the tendon on T2-weighted images, and whether the tendon enhanced after gadolinium administration.
The distance from the tendon involved in the infection to the skin ulcer was measured on a PACS (picture archiving and communication system) workstation (Canon Medical Systems, Irvine, CA), and whether the infected tendon was contiguous to an abscess was noted. Spread of the infection along tendons and the direction of spread (proximal or distal) were noted. In cases of spread of infection along the flexor tendons of the toes, the reviewers also noted whether an abscess was present in the central plantar compartment [12, 13]. In patients with spread of infection along the tendons of the hindfoot, whether the infection had ascended to the lower leg muscle compartment, as described in previous reports [14, 15], was noted. The presence and location of ulcers and osteomyelitis on MR images were also recorded.
After undergoing MR imaging, all patients underwent surgical treatment for the infection by débridement or amputation. All the patients' charts and surgical reports were reviewed by a research fellow who was not analyzing the images. All surgical reports were reviewed to determine whether tendon infection was mentioned as an indication for surgery, whether evidence of tendon infection was documented intraoperatively in the description of the procedure, and whether the surgical procedure was extended or altered because of tendon involvement in the infection or because of peritendinous spread of infection. The research fellow also noted whether infection of the soft tissues was described in the direct vicinity of the tendons presumed to be infected on the basis of MR signal criteria. Bone samples to evaluate for osteomyelitis were collected from all feet. The pathology and microbiology reports for the collected bone samples were also reviewed.
Statistical comparison of the frequency of tendon infection in patients with and those without diabetes mellitus and in patients with and those without osteomyelitis was performed using the chi-square test. Statistical analysis of the frequency of spread of infection along tendons in patients with and those without diabetes mellitus and in patients with and those without osteomyelitis was performed using Fisher's exact test. Before conducting this study, we obtained approval from our hospital's institutional review board to review patients' medical charts and images.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionAPPENDIX 1. MR Imaging...References |
|---|
Evidence of a forefoot infection including the toes was present in 129 of all 159 MR examinations. In 56 of these examinations (43%), evidence of a tendon infection as defined in Appendix 1 was seen. The distribution of tendon infections is summarized in Table 1. These tendon infections in the forefoot were related to ulcerations at the following locations: plantar aspect of the foot (n = 26); medial or lateral aspect of the toes, forefoot, or web space (n = 17); dorsal aspect of the foot (n = 9); or the site of a surgical defect (n = 4) (Fig. 1A,1B,1C).
|
|
|
|
A hindfoot infection was seen in 32 examinations. In 14 (44%) of these 32 examinations, MR evidence of a tendon infection was observed (Fig. 2A,2B,2C,2D and Table 1). All presumed infections of the Achilles tendon (n = 7) were confined to the insertion at the calcaneus (Fig. 3A,3B,3C).
|
|
|
|
|
|
|
MR Evidence of Peritendinous Extension of Infection
Extension of the pedal infection along tendons was observed exclusively in
patients with diabetes. MR evidence of peritendinous spread was observed in
seven patients in the forefoot, including the toes and metatarsophalangeal
joint regions (Fig.
4A,4B,4C,4D),
and in five patients in the hindfoot. Table
2 summarizes the involved tendons, the mean distance of
peritendinous spread of infection, and the observed complications of
peritendinous spread in the forefoot and hindfoot.
|
|
|
|
|
Surgical Correlation
All 68 patients with MR evidence of tendon infection underwent surgical
intervention after a mean of 8.2 days after MR evaluation. The following
procedures were performed: ray amputation (n = 20), toe amputation
(n = 18), débridement (n = 17), transmetatarsal
amputation (n = 6), below-knee amputation (n = 4), foot
amputation (n = 1), Chopart's amputation (n = 1), and
calcanectomy (n = 1).
Evidence of a tendon infection was intraoperatively described in 11 patients. The presence and resection of infected soft tissues adjacent to tendons with MR evidence of infection were documented in 25 patients' surgical records: 18 surgical reports described resection of infected seropurulent tissue, and seven surgical reports described resection of gangrenous, necrotic infection in ulcers with MR evidence of tendon infection. The other 32 surgical reports described only the technical performance of the procedure without mentioning the intraoperative findings. The surgical pathology reports did not describe evidence of tendon infection or destruction.
Eight of the 11 patients with surgically documented tendon infection were treated with amputation, and three had débridement with resection of the involved tendons. All patients with surgically documented gangrenous infection and 14 of the 18 patients with surgically documented infection around tendons with presumed infection had amputation. The remaining four patients had débridement. Nine of the patients with diabetes mellitus who had MR evidence of peritendinous spread of infection underwent amputation because of extensive infection, and three had local débridement.
None of the surgical reports mentioned tendon involvement or peritendinous extension of infection as the cause for surgery. In six surgical reports, extension or adaptation of a surgical procedure related to a tendon infection or peritendinous spread of infection was discussed: resection of the infected soft tissues along a tendon with MR evidence of peritendinous spread was described in two reports; and in a patient with septic tenosynovitis, pus was cleaned out of the infected tendon sheath during débridement of the ulcer. All three central plantar compartment abscesses seen on MR images arising from flexor tendon infection were confirmed at surgery.
The presence of osteomyelitis was confirmed by pathology or microbiology reports in 57 (84%) of the 68 patients with MR evidence of a tendon infection. Eight (73%) of the 11 patients with surgically documented tendon infection and 21 (84%) of the 25 patients with documented presence of peritendinous soft-tissue infection had underlying osteomyelitis.
MR Signal Alterations of Involved Tendons
The signal alterations of all 105 tendons with MR evidence of infection are
summarized in three subgroups in Table
3: surgically documented tendon infection, surgically documented
peritendinous infection, and no surgical correlation. All tendons with
surgically documented infection and surgically documented peritendinous
infection had rim enhancement.
|
In the forefoot, 12 (23%) of 52 involved flexor tendons had no hyperintense rim on T2-weighted images, and 17 (55%) of 31 extensor tendons had a hyperintense rim on T2-weighted images. In the hindfoot, two involved tendons at the medial malleolus and one Achilles tendon lacked a hyperintense rim on T2-weighted images.
Association of Tendon Infection with Diabetes Mellitus or
Osteomyelitis
In our study group, the patients with diabetes mellitus did not have MR
evidence of tendon infection statistically more frequently than patients
without diabetes: fifty-six had diabetes mellitus (43.1%) and 12 did not
(42.8%) (p = 0.85). However, MR evidence of a tendon infection was
statistically more frequent in patients with osteomyelitis: fifty-seven
patients with osteomyelitis (47.5%) and 11 patients without osteomyelitis
(26%) had MR signal changes characteristic of a tendon infection (p =
0.03). Although evidence of spread of infection along tendons was seen only in
patients with diabetes mellitus, this finding was not statistically
significant (p = 0.17).
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionAPPENDIX 1. MR Imaging...References |
|---|
In our study group consisting of patients with advanced pedal infection who all had surgical therapy, nearly half had MR evidence of concomitant tendon infection. Most patients in our study group had diabetes, and all infections were contiguous infections resulting from skin ulcers. The most frequently involved tendons were the tendons of the first and fifth rays and the Achilles tendon, which mirrors the distribution of sites with frequent ulceration as discussed earlier. In the forefoot, nearly two thirds of tendon infections involved the flexor tendons because of the predominant plantar location of forefoot ulceration.
Contrast enhancement of a tendon and the presence of peritendinous fluid signal on T2-weighted images are not specific for the presence of a tendon infection [3, 6]; these findings can also be seen in patients with inflammatory conditions [23, 24], pigmented villonodular synovitis [25], and degenerative disorders of the tendons [26] and in patients who have sustained trauma. However, circular enhancement of a tendon passing through an area of cellulitis from an infected pedal ulcer may be a specific sign of infection [27]. Circular peritendinous contrast enhancement was notably the only MR signal change that was observed in all intraoperatively documented tendon infections in our population. Other signal alterations such as hyperintense signal of the tendon, tendon enhancement, or thickening of the tendon diameter were observed in only one half to one fourth of these tendons. The surgeons probably mentioned tendon infections in the surgical reports only if advanced macroscopic changes of the tendon were seen. This explanation may account for the higher percentage of tendon enhancement, edema, and thickening in this group compared with the other presumed tendon infections. Abnormal hyperintense signal on T2-weighted images, contrast enhancement of the tendon, and thickening may therefore represent findings of advanced tendon infection.
The presence of a peritendinous rim of fluid signal on T2-weighted MR images was not required in our study to establish the diagnosis of a tendon infection because peritendinous fluid signal can be a normal finding around the ankle joint [28] and may relate to the physiology of patients with diabetes [27]. In our patient group, we observed a peritendinous rim of fluid signal in nearly half of the infected extensor tendons of the toes, although these tendons lack sheaths [11] (Fig. 1A,1B,1C). This circular fluid signal may therefore represent paratendinous edema rather than true tenosynovitis in this location. The distal flexor tendons of the toes, on the other hand, are embedded in individual tendon sheaths that stretch from the metatarsal head to the base of the distal phalanx [11]. However, in our population, one fourth of the infections of the flexor tendons of the toes did not have a circular rim of fluid signal. This observation indicates that peritendinous fluid may not always be seen in the setting of infection, perhaps because perforation of the tendon sheath allows fluid leakage, fluid is replaced by solid synovial tissue, or the amount of fluid is too small to detect.
In the hindfoot, all tendons except the Achilles are covered by tendon sheaths as they course behind the medial and lateral malleolus or in front of the ankle joint [11]. In our study group, two infections of the hindfoot did not have a fluid rim around the infected tendons behind the medial malleolus. This observation suggests that, as in the flexor tendons of the toes, this sign of tenosynovitis may not always be present in the hindfoot. All but one Achilles tendon infection had a surrounding rim of fluid signal that may represent local surrounding edema rather than synovitis because the Achilles tendon lacks a tendon sheath [29].
Bacterial infection of foot tendons is discussed in clinical reports almost exclusively in conjunction with proximal spread of infection along the course of these tendons [12, 13]. The tendons that have most frequently been described as involved in peritendinous spread of infection are the flexor tendons of the toes [13, 30]. In a series of 300 patients with major foot infections, 80% had abscesses in the deep plantar compartment, and most of these abscesses originated from peritendinous spread of toe and web space infections [30]. Deep web space infections were reported to extend to the bursae of the lumbrical tendons and to spread along the lumbrical muscles into the central plantar compartment [12]. Infections of the plantar surfaces of the toes were reported to spread along the sheath of the common flexor tendon into the central plantar compartment [12, 13].
In our patient group, only seven MR examinations showed evidence of spread of infection along tendons of the forefoot, and only three of these infections led to typical plantar compartment abscesses located around the flexor tendons (Fig. 3A,3B,3C). A possible explanation for the disparity in frequency of spread of infection along the flexor tendons in our population versus study groups from prior reports is that patients from the earlier studies may have presented with more advanced infections. Many patients in our population were closely monitored and were under ambulatory treatment by a subspecialty physician.
Spread of infection along the flexor tendons of the hindfoot into the lower leg is frequently cited as a complication of deep pedal infection [14, 15]. Direct injections of dyes and contrast agents into the deep plantar compartments of cadaveric feet have confirmed extension of injected substances along the flexor tendon sheaths around the medial malleolus into the lower leg [15, 31]. In our study group, however, infectious spread along hindfoot tendons always originated from hindfoot ulcers and never from a deep pedal infection. Spread of infection from the plantar pedal compartments of the foot into the lower leg was not observed in our study.
Review of the surgical reports in our study revealed that tendon infection influenced the surgical procedure infrequently. More than 80% of our patients with MR evidence of a tendon infection had biopsy-proven underlying osteomyelitis that, in most patients, was the reason to perform the surgical intervention. In most patients, the intervention consisted of amputation. Adaptation or extension of a surgical procedure because of a tendon infection was performed in fewer than 10% of our patients with evidence of a tendon infection. This finding is in agreement with our foot surgeon's experience. However, careful evaluation of potential tendon infections during MR analysis may reveal peritendinous spread of infection, plantar compartment abscesses, or advanced tenosynovitis, which may influence the surgical procedure.
MR signal changes compatible with tendon infection were observed in nearly equal proportions of patients with and those without diabetes. However, we found a statistically significant correlation between the presence of osteomyelitis and the presence of a tendon infection (p = 0.03). This correlation is best explained by the fact that both tendon involvement and osteomyelitis are complications of advanced deep infections. Although spread of infection along tendons occurred exclusively in patients with diabetes, this proportion did not reach statistical significance for our study population.
Some limitations of our study should be mentioned: The selection of patients who needed surgical treatment after MR imaging may have resulted in a higher frequency of tendon infections and a higher frequency of peritendinous spread in our study group than in a randomized patient group. Tendon infections were diagnosed according to MR criteria, which may have led to the inclusion of tendons with aseptic peritendinous inflammation. Surgical proof could be obtained in only a minority of patients because of the retrospective nature of this study. Surgical documentation relied on inspection and palpation of the involved tendon or description of peritendinous soft-tissue infection without culture proof. Review of the surgical reports may have resulted in a smaller number of documented tendon infections and of documented paratendinous spread of infection and in an underestimation of the importance of tendon involvement concerning the surgical procedure because the reports did not specifically focus on tendon involvement.
We conclude that nearly half of the patients who need surgical treatment for pedal infection have MR evidence of a tendon infection that most frequently involves the flexor tendons of the forefoot. Evidence of spread of infection along tendons is seen in a minority of patients and relevant extension of infection is seldom observed. However, MR evidence of tendon infection may be useful preoperative information that could potentially alter the surgical treatment in a minority of patients.
APPENDIX 1. MR Imaging Criteria Used to Describe Manifestations and
Complications of Pedal Infections
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionAPPENDIX 1. MR Imaging...References |
|---|
|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionAPPENDIX 1. MR Imaging...References |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  P L Tan and J Teh MRI of the diabetic foot: differentiation of infection from neuropathic change Br. J. Radiol., November 1, 2007; 80(959): 939 - 948. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. L. Bloem, T. J. Mosher, M. R. Schmid, M. Zanetti, H. P. Ledermann, W. B. Morrison, and M. E. Schweitzer Dynamic Gadolinium-enhanced MR Imaging in Bone Marrow Disorders [letter] * Dr Mosher responds: * Drs Schmid and Zanetti respond: * Dr Ledermann and colleagues respond: Radiology, April 1, 2003; 227(1): 303 - 305. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
