| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1
Department of Radiology, Thomas Jefferson University Hospital, 111 S. 11th
St., #3390 Gibbon, Philadelphia, PA 19107.
2
Present address: FMH diagnostische Radiologie, Oberer Batterieweg 57, 4059
Basel, Switzerland.
Received April 26, 2001;
accepted after revision August 29, 2001.
Presented at the annual meeting of the American Roentgen Ray Society,
Seattle, April-May 2001.
Abstract
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
MATERIALS AND METHODS. We reviewed 115 contrast-enhanced 1.5-T MR examinations of the foot in 41 women and 74 men with a mean age of 58.4 years who had undergone bone biopsy or surgery for suspected osteomyelitis. Presence of inflammation (contrast enhancement, fat signal intensity loss on T1-weighted images, and high signal intensity on T2-weighted images) was noted by two musculoskeletal radiologists in the following foot compartments: toes, medial, central, lateral, interosseous, dorsal, hindfoot, malleoli, and lower leg. Proximal and distal extension of soft-tissue inflammation was analyzed. The compartment closest to the ulcer that showed MR signs of direct contiguous infection was designated the primarily infected compartment.
RESULTS. Spread of inflammation across fascial planes into neighboring compartments originated from the following primary compartments: medial (3/10, 30%), central (7/16, 44%), and lateral (16/20, 80%). Spread from the hindfoot and malleoli into adjacent compartments was seen in only 7% of such cases (2/24). Inflammation from toe infections spread in 34% of cases to forefoot compartments (15/44). Inflammation from forefoot or toe infections spread in 4.5% of cases to the midfoot and in 2% of cases to the hindfoot; ascension into the calf was rare (1% of cases). Spread of inflammation into neighboring compartments was not correlated with the presence of diabetes (p = 0.81) or with osteomyelitis (p = 0.34).
CONCLUSION. Soft-tissue inflammation of the forefoot tends to spread into neighboring compartments, with little respect for fascial planes. Hindfoot inflammation tends to stay confined. Spread from the foot to the lower leg is rare.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
|
MR imaging is often used to evaluate the extent of pedal infection [8,9,10,11], because it allows precise demarcation of infected tissue [8,9,10]. The purpose of our study was to evaluate the compartmental distribution of soft-tissue inflammation in pedal infection in a large group of patients using MR imaging. In particular, we wanted to describe patterns of soft-tissue inflammation and to determine whether spread of inflammation is contained by fascial planes.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
MR imaging was performed with a 1.5-T unit (Signa; General Electric Medical Systems, Milwaukee, WI). An extremity coil was used for 112 feet in 112 patients (field of view, 14-20 cm), and a head coil was used for three patients who underwent imaging of both feet (field of view, 16-20 cm). In these three patients, only one foot each was included in this study. Images of all feet were obtained in at least two, and almost always three, orthogonal planes.
T1-weighted spin-echo images were obtained usually with two signals averaged (TR range/TE range, 400-750/10-20; echo-train length, 8; matrix, 256 x 192 or 256 x 256). T2-weighted images were acquired using a fast spin-echo technique, with 2 signals averaged (TR range/TEeff range, 2000-7800/75-108; matrix, 256 x 128 or 256 x 192). Gadolinium-enhanced T1-weighted images were obtained using spin-echo sequences with fat suppression in 11 patients and fast multiplane spoiled gradient-recalled sequences with fat suppression in 104 patients. T1-weighted spin-echo sequences were obtained with 2 signals averaged (TR range/TE range, 450-800/10-20; matrix, 256 x 128 or 256 x 192). Fast multiplane spoiled gradient-recalled sequences were obtained using TR/TE of 250/2.1 and a flip angle of 90°. Imaging began after IV administration of gadopentetate dimeglumine (Magnevist; Berlex Laboratories, Wayne, NJ) at a dose of 0.1 mmol/kg of body weight. Gadolinium-enhanced images were available for all feet. For all T2-weighted and gadolinium-enhanced T1-weighted sequences, fat suppression was accomplished using selective presaturation of lipid resonant frequencies. Fast spin-echo short tau inversion recovery (STIR) images were obtained with TR range/TEeff range of 2000-3000/20-78, inversion time of 150-160 msec, echo-train length of 8, and matrix of 256 x 128 or 256 x 192. Fast spin-echo STIR images were available for 108 feet.
Associated medical conditions in the 115 patients, which were determined by review of their medical records, included diabetes mellitus (n = 91), paraplegia (n = 6), prior trauma (n = 3), peripheral vascular disease (n = 5), poor hygiene associated with mental disorder (n = 4), alcohol abuse (n = 2), severely ingrown toenail (n = 1), frostbite (n = 1), vascular embolism (n = 1), and sickle cell disease (n = 1).
A spreadsheet was designed to facilitate data acquisition of compartmental involvement and spread of inflammation. This form contained two schematic coronal sections of a foot at the level of the forefoot and midfoot (Fig. 1A,1B). The borders of the medial, central, lateral, interosseous, and dorsal compartments were depicted on these coronal sections, as described in previous reports [2, 3, 5, 6, 12,13,14]. In addition, two schematic drawings of a foot and ankle joint in anteroposterior and lateral projections were included on the spreadsheet to facilitate recording of the proximal and distal extent of inflammation. To define proximal and distal extent of inflammatory signal changes, the foot and ankle were divided into six anatomic regions: toes, forefoot (from the metatarsophalangeal joint to the Lisfranc's joint), midfoot (from the Lisfranc's joint to the Chopart's joint), hindfoot (from the Chopart's joint to the ankle joint), malleoli (the medial and lateral malleolus), and lower leg (from the ankle up).
Two musculoskeletal radiologists reviewed together all 115 MR examinations. The reviewers noted on each examination the presence and location of the ulcer as the starting point of infection. Diagnosis of a skin ulcer relied on recognition of a soft-tissue defect or an interruption of the skin line on any imaging sequence or plane [15]. Then the primarily infected compartment was determined. Soft-tissue infection and inflammation was defined on MR images as contrast enhancement, fat signal intensity loss on T1-weighted images, and hyperintense signal on T2-weighted images compared with that of muscle, as described by prior authors [15,16,17,18,19]. The primarily infected compartment was defined as the compartment closest to the ulcer that also displayed MR imaging signs of direct contiguous infection from the ulcer. The reviewers then noted whether evidence was found of contiguous spread of inflammation originating from the primarily infected compartment into adjacent compartments. Spread of inflammation was defined as fat signal intensity loss on T1-weighted images, hyperintense signal on T2-weighted images, and contrast enhancement extending contiguously from the primarily involved compartment across fascial borders into neighboring compartments. The reviewers also noted whether abscesses were present. MR imaging criteria for the diagnosis of an abscess included isointense or hypointense signal compared with that of muscle tissue on T1-weighted images, fluid equivalent signal on T2-weighted images, and rim enhancement on gadolinium-enhanced T1-weighted images [20, 21]. On the schematic drawings on the spreadsheet for each patient, the reviewers noted the location of the ulcer, the primarily infected compartment, and the observed routes of spread of soft-tissue inflammation into neighboring compartments. In patients with MR evidence of an abscess, the reviewers determined potential extension of the abscess into adjacent compartments. Involvement of the dorsal compartment as the primarily infected compartment and as the secondarily involved compartment required evidence of inflammation of the deep subaponeurotic portion [2] of the dorsal compartment. Spread of inflammation from the medial, lateral, or plantar subcutaneous compartment to the subcutaneous dorsal compartment was not considered as spread between compartments, because no fascial borders divide the dorsal, plantar, medial, or lateral, subcutaneous compartments. The distal and proximal extension of soft-tissue inflammation was also noted on the spreadsheet.
MR imaging criteria for the diagnosis of osteomyelitis were based on those described in previous studies [10, 11, 12], including focally decreased marrow signal intensity on T1-weighted images, focally increased signal intensity on fat-suppressed T2-weighted and fast spin-echo STIR images, and focal marrow enhancement on gadolinium-enhanced fat-suppressed T1-weighted images. The patients' charts were reviewed to determine whether osteomyelitis was proven or excluded by biopsy or by intraoperative histology. The location of the infection as determined by MR analysis was compared with the location as described in the surgical reports and the patients' charts.
A chi-square test was performed to determine whether spread of inflammation across fascial planes occurred statistically more frequently in patients with osteomyelitis than in patients without bone infection. Similarly, Fisher's exact test was performed to determine whether spread of inflammation across fascial planes occurred more frequently in diabetic patients than in patients without diabetes mellitus. A t test was performed to compare the number of secondarily inflamed compartments in patients with and without osteomyelitis and diabetes mellitus. A p value of 0.05 was considered to indicate a significant difference.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Of the 44 infections which originated in the toes, proximal spread of inflammation into the forefoot was found in 15 cases (34%). Spread was from the toes into the central compartment in five (11%) of the cases, interosseous compartment in six (14%), medial compartment in five (11%), lateral compartment in four (9%), and dorsal compartment in seven (16%). Two toe infections spread to adjacent toes (4%). In eight toe infections (18%), spread occurred into two or more forefoot compartments.
Forty-five infections started in the forefoot and two in the midfoot. One infection in the midfoot originated on the medial plantar surface of the foot in the area of the medial cuneiform. The other midfoot infection originated from an ulcer on the lateral aspect of the midfoot in the area of the cuboid.
The medial compartment was the primarily infected site in 10 patients, and the most frequently observed pattern of spread of inflammation was distally into the toes (n = 6, 60%). Spread of soft-tissue inflammation across fascial planes into neighboring compartments occurred in three examinations (30%) and involved the central compartment (Figs. 2C and 2D) (n = 2, 20%), the interosseous compartment (n = 2, 20%), and dorsal compartment (n = 1, 10%).
|
|
The central compartment was primarily involved in 16 examinations. The most frequently observed pattern of inflammatory spread was distally into the toes (n = 12, 75%). Spread into neighboring compartments of the forefoot occurred in seven feet (44%) and involved the interosseous compartment (n = 6, 37%), dorsal compartment (n = 3, 19%), and lateral compartment (n = 1, 6%).
Primary involvement of the lateral compartment (Fig. 3A,3B,3C) occurred in 20 feet, and inflammation most frequently spread into the toes (n = 13, 65%). Spread into neighboring compartments was seen in 16 examinations (80%) and involved the dorsal compartment (n = 10, 50%), central compartment (n = 9, 45%), interosseous compartment (n = 10, 50%), medial compartment (n = 1, 5%), hindfoot (n = 2, 10%), and lower leg (n = 1, 5%).
|
|
|
Primary infection of the dorsal compartment occurred in one patient, in whom spread of soft-tissue inflammation extended into the interosseous compartment (n = 1, 100%) and the medial compartment (n = 1, 100%).
Spread of inflammation from the central compartment into the bordering medial and lateral compartments was seen in only one of 16 patients (6%), whereas spread into the central compartment from the medial and lateral compartments was seen in 20% (n = 2) and 45% (n = 9), respectively.
Distal spread of soft-tissue inflammation from the forefoot or midfoot into
the toes was seen in 32 (68%) of 47 infections. Proximal spread of
inflammation in forefoot and midfoot infections was infrequently seen. Only
four (4.5%) of 89 feet with forefoot infections (including toes) had extension
of soft-tissue inflammation into the midfoot: one had an acute streptococcal
infection, and three had an extensive infection of the lateral compartment.
Proximal spread of inflammation from the forefoot into the hindfoot was seen
in two of these feet (2%) with lateral compartment infection. Only one
infection of the forefoot or midfoot (1%) spread to the lower leg (from the
lateral compartment in the
forefoot).
,
|
|
The hindfoot was the primarily infected site in 21 feet (Fig. 4A,4B,4C,4D). In this group, heel ulcers were the focus of infection in 19 feet, one had a wound dorsally over the hindfoot, and one had an ulcer proximal to the insertion of the Achilles tendon. Spread into other compartments was seen on only one MR examination, which revealed massive infection from a heel ulceration, with proximal spread into the lower leg, and distal extension into the central and lateral compartments.
|
|
|
|
Three infections started around the medial (n = 1) and lateral (n = 2) malleoli, and one of these lateral malleoli infections spread to the hindfoot and lower leg.
After MR evaluation, bone biopsy was performed in five patients, above knee amputation in three, below knee amputation in 11, transmetatarsal amputation in nine, ray amputation in 22, toe amputation in 39, and débridement in 26. Most primarily infected compartments (n = 87; 75.7%) had osteomyelitis proven by histology or bacteriology. In the remaining 28 primarily infected compartments without osteomyelitis, presence of infection was confirmed by surgical reports (n = 25), positive cultures from intraoperative swabs (n = 19), and histology reports (n = 2).
Spread of soft-tissue inflammation across fascial planes of the forefoot occurred in 22 patients with osteomyelitis (47%) and in five patients without osteomyelitis (42%) (p = 0.34). Spread of inflammation occurred into a mean of 0.7 compartments in patients without osteomyelitis and into a mean of 1.2 compartments in patients with osteomyelitis (p = 0.18).
Most patients with forefoot and midfoot infections had diabetes mellitus (38/47, 81%). In the entire study group, 79% (91/115) of patients were diabetic. Spread of inflammation across compartmental borders was seen in 21 patients with diabetes (55%) and six patients without diabetes (67%) (p = 0.81). Spread of inflammation occurred into a mean of 0.9 compartments in patients without diabetes mellitus and into a mean of 1.1 compartments in patients with diabetes mellitus (p = 0.7).
Thirteen patients had evidence of an abscess in primarily infected compartments, including the hindfoot (n = 4), central compartment (n = 3), lateral compartment (n = 3), toes (n = 2), and medial compartment (n = 1). All 13 patients had concomitant osteomyelitis in their primary involved compartments. Abscess extension into adjacent compartments was observed in six patients: two abscesses of the lateral compartment extended into the interosseous and dorsal compartments and one into the fifth toe. One abscess of the central compartment extended into the lateral and interosseous compartment. One collection of the medial compartment extended into the interosseous compartment, and one abscess of the fifth toe extended into the lateral compartment. All four abscesses in the hindfoot remained localized without extension into adjacent compartments.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
In addition to the plantar compartments, most authors acknowledge the existence of interosseous compartments [2, 3, 5, 14, 23] and some also describe dorsal compartments [2, 12]. The interosseous compartment is subdivided into four spaces containing the corresponding interosseous muscles. This compartment is limited on the plantar side by the thin interosseous fascia and dorsally by the dorsal interossei aponeurosis [14].
The dorsal foot space is divided into a subcutaneous and subaponeurotic portion [2]. The subaponeurotic portion contains the extensor tendons of the toes and is enclosed dorsally by the deep foot fascia and on the plantar aspect by the dorsal interosseous membrane [2].
Because spread infection in the hand is thought to depend on compartmental anatomy, historically, description of foot compartments was targeted to define similar pathways for the spread of infection [2, 3, 6, 12]. However, important differences exist between patients with hand infections and patients with foot infections, because patients with hand infections are usually younger and are generally otherwise healthy. Most pedal infections, however, result from skin ulceration [24] in patients with predisposing conditions such as diabetes mellitus and peripheral vascular disease that favor propagation and chronicity of soft-tissue infection [25, 26]. Cross-sectional evaluation of compartmental involvement in pedal infection with systematic analysis of the findings has been reported only once [13]. However, that study used fluid collections on unenhanced MR images to define spread of infection, and the population studied was quite small, including only 11 patients.
Despite concern for proximal spread, our study found that, by far, the most frequently observed pattern of spread of inflammation in forefoot infection was distally into the adjacent toes. Spread from the forefoot into the toes does not cross fascial planes and, therefore, does not represent spread between closed compartments. Because almost all of these infections originated from plantar ulcers around the metatarsal heads, spread may have occurred along the adjacent flexor tendons, digital nerves, and lumbricals into the toes [7]. The hypothesis of distal infectious spread along tendons and lumbrical muscles is supported by prior studies using injection of gelatin into the superficial portions of the central plantar compartment: Distal extension of gelatin into the peritendinous interspaces close to the base of the toes was seen in all injections in cadaveric feet [2]. Distal filling of the lumbrical sheaths to the bases of toes has also been demonstrated by injections of the central compartment [6]. However, proximal spread of soft-tissue inflammation from the toes into the forefoot was seen in one third of all toe infections in our series. This pattern of spread is well known in the orthopedic and surgical literature because it may lead to a plantar abscess [7, 27].
In our study, nearly one third of infections of the medial compartment, nearly half of infections of the central compartment, and four fifths of infections of the lateral compartment had evidence of inflammatory spread across their fascial boundaries into neighboring compartments. Instead of extending proximally within the boundaries of the primarily infected compartment, forefoot and midfoot infections tended to spread into adjacent compartments, seemingly without regard to fascial barriers (Figs. 2C, 2D, 3B, and 3C). Concordant to the observed spread of soft-tissue inflammation, nearly half of the abscesses extended into adjacent compartments.
Extension of inflammation was most frequently observed in infections of the lateral compartment, in which only one fifth of infections did not spread into neighboring compartments. Frequent extension of soft-tissue inflammation and abscess formation across the fascial borders of the lateral compartment may be explained by its small size [2, 6]. Even a limited infection of this compartment soon abuts its fascial boundaries. The fascia of this compartment may furthermore be weaker than in the other plantar compartments, because extravasation from direct injections into the lateral compartment was observed more frequently here than in the other plantar compartments [6].
Several pathophysiologic mechanisms may affect the integrity of fascial boundaries in pedal infections. Infected tissues in the critically ischemic extremity may proceed to focal necrosis because the metabolic requirements to heal the infection are far greater than those required to maintain normal tissue viability [26, 28]. Confirmatory evidence of this hypothesis is that vertical invasion of soft-tissue infection into the perifascial plane occurs more frequently in the ischemic extremity than elsewhere [26]. Chronic inflammation may lead to thrombotic interruption of the microvasculature, thus causing further cell death and damage to the integrity of host tissue [29]. Necrotic tissue offers an ideal environment for bacterial proliferation and promotes a sustained inflammatory reaction with damage to the surrounding structures by proinflammatory mediators and proteinases [25]. Bacterial enzymes degrade the integrity of host tissues [29]. Collagenase production by Str. pyogenes, S. aureus, and P. aeruginosa degrade the fundamental collagen structure that constitutes the "skeleton" of many tissues [30].
We observed inflammatory spread into the central compartment in one fifth of medial compartment infections and nearly half of lateral compartment infections and nearly half of lateral compartment infections. Frequent involvement of the central compartment by infection of the adjacent medial and lateral compartments has been noted in earlier reports [12, 13] and may be caused, in part, by contiguous spread along anatomically preformed pathways that violate the medial and lateral intermuscular septum [12]. Five musculotendinous structures traverse the lateral intermuscular septum (flexor digitorum brevis and longus to fifth toe, lumbrical and interosseous to fifth toe, and peroneus longus), and four musculotendinous structures traverse the medial intermuscular septum (adductor hallucis, flexor hallucis longus, peroneus longus, and tibialis posterior). It has also been suggested that frequent involvement of the central compartment may be caused by direct extension of infection rather than by spread across fascial planes [13], because the large central compartment often abuts infectious foci around the first and fifth metatarsal head. However, spread from the central compartment into the adjacent lateral and medial compartment was observed in only one patient in our study group. A possible explanation for rare spread to the adjacent medial and lateral compartments in central compartment infections is the large size of the central compartment [13].
Earlier reports emphasize the possibility of proximal spread of infection along preformed channels. In particular, potential ascension of infection into the calf from the central compartment along flexor tendons was described by many authors after injection experiments [2,3,4,5,6]. In our 115 patients, however, proximal spread from the forefoot and midfoot was rarely seen: only four patients with forefoot infection had extension of soft-tissue inflammation into the midfoot. Proximal extension from the forefoot into the hindfoot was seen in only two patients, both with very extensive infections. Ascension of a forefoot infection to the calf occurred in only one patient, who had an advanced infection of the lateral compartment. These observations may relate to selection bias, and we cannot exclude the possibility that proximal spread may occur with late disease. In our study, most infections of the hindfoot (Fig. 4A,4B,4C,4D), somewhat different from proximal infections, tended to be confined to their site of origin, with only rare spread to the foot and calf.
Underlying diabetes mellitus did not result in increased occurrence of spread, nor did it lead to involvement of considerably more secondary compartments. Unfortunately, statistical analysis in the forefoot and midfoot included only nine patients without diabetes mellitus, which somewhat limits the value of this analysis. Although infections with underlying osteomyelitis had inflammatory spread into more compartments than infections without osteomyelitis, no statistical significance was reached.
Some limitations apply to our study: A selection of advanced infections was performed because we included only those patients who had subsequent bone biopsy or surgery for suspected osteomyelitis. We used this population to specifically define the population with infection. This definition may have resulted in findings of higher frequency of spread of inflammation than would have been found in a random selection of patients. The presence and extent of soft-tissue inflammation was determined by MR criteria without direct intraoperative correlation, because the surgical reports did not specifically mention the extent of soft-tissue involvement. Because presence of infection was proven in all primarily involved compartments, corresponding signal alterations were called "infection" in all primarily involved compartments. Spread of MR signal alterations into neighboring compartments was called "inflammation" rather than "infection," because the presence of infection was not proven in these secondarily involved compartments.
We conclude that soft-tissue inflammation in forefoot infection is not confined by fascial planes but spreads into adjacent compartments, whereas hindfoot infection tends to stay confined with only rare spread to other compartments. Inflammation in forefoot infections most frequently spreads distally into adjacent toes. Proximal spread to the midfoot and hindfoot occurs seldom and ascension to the claf is rare.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  B. Veres, B. Radnai, F. Gallyas Jr., G. Varbiro, Z. Berente, E. Osz, and B. Sumegi Regulation of Kinase Cascades and Transcription Factors by a Poly(ADP-Ribose) Polymerase-1 Inhibitor, 4-Hydroxyquinazoline, in Lipopolysaccharide-Induced Inflammation in Mice J. Pharmacol. Exp. Ther., July 1, 2004; 310(1): 247 - 255. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
