1. Top of page
  2. Abstract
  3. Introduction


To determine whether targeted ultrasonographic (US) imaging of the fifth metatarsophalangeal (MTP) joint, compared with radiographs, could aid in the early diagnosis of rheumatoid arthritis (RA) by identifying erosions sooner in early inflammatory arthritis. Radiographic erosion in RA is a late indication of poor prognosis. The earlier detection of erosion may facilitate the timely initiation of disease-modifying antirheumatic drug therapy, particularly in patients with undifferentiated synovitis.


Patients presenting with synovitis for the first time were invited to participate. Each patient underwent laboratory tests, radiographs of the hands and feet, and US imaging of both fifth MTP joints.


Thirty patients (22 women) took part in the study. Seventeen patients (57%) had RA, and 13 (43%) had undifferentiated arthritis (UA). The mean ± SD time taken to scan both fifth MTP joints was 10.9 ± 4.4 minutes. Ten patients (33%) had US evidence of synovitis associated with a positive power Doppler (PD) signal (P = 0.04). Seven patients (23%) had radiographic erosions of the fifth MTP joint, and 17 patients (57%) had US evidence of fifth MTP joint erosions (P = 0.01). A positive PD signal at the fifth MTP joint was seen in 9 of 17 patients with RA and 1 of 13 patients with UA (P = 0.02). Patients with a definite diagnosis of RA were more likely to have fifth MTP joint erosions (11 [65%] of 17) compared with UA (6 [46%] of 13).


Targeted US is a rapid and useful tool in detecting erosive disease in early inflammatory arthritis. It gives a better indication of disease severity and prognosis compared with routinely available laboratory tests, even in the absence of a definite diagnosis.


  1. Top of page
  2. Abstract
  3. Introduction

Rheumatoid arthritis (RA) is a systemic inflammatory polyarthropathy characterized by progressive joint damage and nonarticular complications such as osteoporosis (1), accelerated atherosclerosis, and increased risk of malignancy (2). The earliest appropriate initiation of treatment after disease onset offers the best chance of permanent remission and a normal lifespan. In this regard, early diagnosis is essential but is frequently hampered by the lack of sufficient clinical evidence early in the disease course (3).

Joint erosions identified by conventional radiography are late findings indicating a poor prognosis (4). More sensitive diagnostic tools for the early detection of joint damage include magnetic resonance imaging (MRI) and ultrasound (US) (5, 6). MRI provides excellent detail for articular defects but is expensive and less accessible to rheumatologists compared with US, which can be used at the bedside or in the clinic (7).

One of the first joints to demonstrate abnormalities in patients with RA is the fifth metatarsophalangeal (MTP) joint (8). This study was undertaken to determine if targeted US imaging of the fifth MTP joint could help in the early diagnosis of RA by identifying erosions sooner than conventional radiography in patients with early inflammatory arthritis.

Patients and Methods


Patients with a new diagnosis of inflammatory arthritis seen at rheumatology outpatient clinics were invited to participate in the study. Inflammatory arthritis was defined as early morning joint stiffness lasting >30 minutes and associated with joint pain and swelling, and synovitis in at least 1 joint on clinical examination. RA was diagnosed according to the criteria of the American College of Rheumatology (ACR; formerly the American Rheumatism Association) (9).

Each patient had conventional radiographs taken of both hands and feet in posteroanterior views. These radiographs were read by a group of radiologists different from those performing the US assessments, which took place within 4 weeks of the radiographic assessment of the hands and feet. Each scan was performed by the same radiologist (SM), who was assisted by a radiology fellow (PB). Both were blinded to the radiographic results and the underlying rheumatic diagnosis.


US imaging was performed on a Philips HDI 5000 scanner (Philips, Eindhoven, The Netherlands) using a high frequency linear (CL 15–7-MHz) hockey stick transducer. The frequency was set automatically, and patients were scanned using the hand/foot musculoskeletal setting. This setting uses a survey frame rate and is set to compound imaging (sonoCT); the preset is optimized for imaging superficial structures. One variable focal zone was employed with a narrow field of view. All color imaging was performed with power Doppler (PD). The PD values were set automatically by the machine and optimized for musculoskeletal imaging. The patients were scanned with identical preset values. The focal zone was adjusted as required. The scans were performed in a warm water bath where the entire joint was submerged, which enabled easy access to all the joint surfaces and ensured a high standard of image quality (10). US images were obtained at the fifth MTP joints from 3 different angles. The first angle was from the dorsal aspect of the foot, the second was from the lateral approach, and the third was from the plantar surface of the foot. The probe was passed over the joint at these 3 positions, which represent the available joint surface that is accessible to an US probe, and at each position the joint was examined in the longitudinal and transverse plane. The 3 angles gave almost full coverage of the joint. The time taken to complete each US assessment was recorded.

The semiquantitative scoring system and the definitions of US abnormalities (erosions, synovitis, and PD signal), originally outlined by Szkudlarek et al (11), were applied in this study. Erosions were defined as defects in the bone adjacent to the joint. A score of ≥2 on a scale of 0–3 was required for a defect to be categorized as an erosion (where 0 = regular bone surface, 1 = defect in the bone surface in 1 plane only, 2 = bone surface defect visualized in 2 planes, and 3 = extensive bone destruction).

Age, sex, symptom duration, erythrocyte sedimentation rate (ESR), C-reactive protein (CRP) level, rheumatoid factor (RF), anti–cyclic citrullinated peptide (anti-CCP) antibody status, Disease Activity Score in 28 joints (DAS28), modified Health Assessment Questionnaire (HAQ), use and duration of disease-modifying antirheumatic drugs (DMARDs), and corticosteroid exposure were all recorded at baseline. At the time of US imaging, 5 patients had begun DMARD therapy, and 25 patients had never taken DMARDs. Of those patients taking DMARDS, the mean time of use was ∼2 weeks and, therefore, was unlikely to have influenced imaging outcomes. However, the use of DMARDs was included as a study variable in the statistical analysis. The use of corticosteroids and nonsteroidal antiinflammatory drugs was permitted. Informed consent was provided by all patients included in the study, and ethical approval was granted by the St. James's Hospital and Federated Dublin Voluntary Hospitals Joint Research Ethics Committee.

Statistical analysis.

Data were recorded and analysed using SPSS statistical software, version 14.0 (SPSS, Chicago, IL). Associations between categorical variables were assessed using chi-square and Fisher's exact tests. Associations between means of parametric data were assessed using Student's t-test.


The baseline characteristics of all the patients are outlined in Table 1. The majority of patients (66%) had joint symptoms for <12 months, and 22 patients (73%) had an articular abnormality detected by US evaluation. The mean ± SD time taken to perform a US of both fifth MTP joints was 10.9 ± 4.4 minutes, which included the time taken to document the findings. Details of the radiographic and US abnormalities detected are outlined in Table 2.

Table 1. Baseline characteristics of the patients (n = 30)
 No. (%)Mean ± SD
Age, years 55.2 ± 17.9
Female22 (73.3) 
Diagnosis at presentation  
 Rheumatoid arthritis17 (56.6) 
 Undifferentiated arthritis13 (43.4) 
 Oral, mg11 (36.6)3.6 ± 5.7
 Intramuscular, mg2 (6.7)80 ± 0
Symptom duration, months 15.2 ± 15.0
Table 2. Imaging, clinical, and serologic characteristics for patients with early RA and UA*
 RA (n = 17)UA (n = 13)Total (n = 30)P
  • *

    Values are the number (percentage) unless indicated otherwise. RA = rheumatoid arthritis; UA = undifferentiated arthritis; MTP = metatarsophalangeal; US = ultrasonographic; PD = power Doppler; RF = rheumatoid factor; anti-CCP = anti–cyclic citrullinated peptide; DAS28 = Disease Activity Score in 28 joints; HAQ = Health Assessment Questionnaire; ESR = erythrocyte sedimentation rate.

  • Based on differences between RA and UA.

  • 95% confidence interval (95% CI) 0.5, 2.4.

  • §

    95% CI −0.2, 1.1.

  • 95% CI 10.3, 34.9.

Radiographic erosions fifth MTP joint7 (41.2)0 (0)7 (23.3)0.01
US erosions fifth MTP joint11 (64.7)6 (46.2)17 (56.7)0.31
US synovitis fifth MTP joint8 (47.1)2 (15.4)10 (33.3)0.12
PD positive fifth MTP joint9 (52.9)1 (7.7)10 (33.3)0.02
RF positive12/15 (80.0)2 (15.4)14/28 (50.0)< 0.01
Anti-CCP antibody positive11/15 (73.3)3 (23.1)14/28 (50.0)< 0.01
DAS28, mean ± SD5.4 ± 1.33.9 ± 1.14.8 ± 1.4< 0.01
HAQ, mean ± SD1.5 ± 0.91.1 ± 0.81.3 ± 0.90.17§
ESR, mean ± SD mm/hour45.1 ± 21.522.1 ± 10.936.1 ± 21.2< 0.01
Radiographic assessment.

Seven (23.3%) of 30 patients had evidence of fifth MTP joint erosions with conventional radiography. Three patients had erosions of the right fifth MTP joint only, 3 had erosions of the left fifth MTP joint only, and 1 patient had bilateral erosive changes. All 7 of these patients fulfilled the ACR criteria for the diagnosis of RA (9). None of the patients with undifferentiated arthritis (UA) demonstrated radiographic abnormalities. Radiographic erosions of the fifth MTP joint were associated with a higher mean ± SD ESR (31.5 ± 17.7 mm/hour [with erosions] compared with 4.7 ± 2.7 mm/hour [no erosions]; P = 0.047) and CRP levels (41.0 ± 31.9 mg/dl [with erosions] compared with 15.1 ± 16.2 mg/dl [no erosions]; P = 0.015). There was no association with age, sex, HAQ score, DAS28, duration of symptoms, RF, anti-CCP antibody, or use of corticosteroids or DMARDs.

Ultrasound erosion.

Seventeen patients had evidence of fifth MTP joint erosions detected by US. Of these 17 patients, 9 had bilateral erosions. There was a significant difference between radiographic (n = 7) and US (n = 17) detection of fifth MTP joint erosions (P = 0.01). Ten patients had erosions demonstrated on US despite an absence of such change on radiographs. All 7 patients with erosions of the fifth MTP joint on radiographs showed abnormalities on US. Compared with radiographs (P = 0.04), US revealed evidence of joint damage and, therefore, a poorer prognosis in a further 23% of the RA cohort. US also revealed erosions in more than 46% of the UA cohort where none had been identified by radiography.

Synovial hypertrophy.

Ten of the patients (33%) had synovial hypertrophy detected by US (2 patients had bilateral fifth MTP joint synovial thickening, 6 had thickening on the right side only, and 2 had thickening on the left side only). There was a significant association between the presence of US synovitis at the fifth MTP joint and radiographic erosions at the same site (P = 0.03). Seven patients had radiographic erosions of the fifth MTP joint, 5 of those patients also had US synovial thickening at the same site, and 7 (41%) of 17 patients with US erosions had US synovial thickening at the same site (P = 0.44).

Power Doppler.

A PD signal at the fifth MTP joint was present in 33% of the patients (5 on the right fifth MTP joint only, 2 on the left fifth MTP joint only, and 3 had bilateral signals). A PD signal at the fifth MTP joint was significantly associated with synovial thickening at the same site (P = 0.04), where 6 of 10 patients with synovial thickening had a PD signal. There was a significant association between radiographic erosions and the presence of a PD signal at the fifth MTP joint (P = 0.03). Nine (53%) of 17 patients with fifth MTP joint erosions on US had a detectable PD signal (P = 0.12). There was no association between US erosions, synovial hypertrophy, or PD signals and ESR, CRP level, RF, anti-CCP antibody, age, sex, HAQ, DAS28, duration of symptoms, or use of corticosteroids or DMARDs.


The presence of radiographic erosions and a PD signal was significantly higher in patients with a diagnosis of early RA compared with those with UA (Table 2). Within 12 months of presentation, 3 patients with UA were subsequently diagnosed with RA based on fulfillment of the ACR criteria for RA (9). Of these 3 patients, none had radiographic erosions at baseline; however, 1 patient had fifth MTP joint erosions detected by US in the absence of synovial hypertrophy or a PD signal. Although not statistically significant, US erosions and synovial hypertrophy were more frequent in patients with early RA compared with those with UA.


The results of this study indicate that US examination of the fifth MTP joint in patients who present with early inflammatory arthritis aids in the identification of those with a poor prognosis and has considerable advantages over conventional radiography in the detection of early erosive disease.

We selected the fifth MTP joint for study since it has previously been identified as an area of early joint damage in patients with RA (6, 8, 12). Although the reasons why this joint appears to be involved at an early stage of the erosive process remain speculative, it has been suggested that this area may be prone to early subluxation because of its location (8), therefore subjecting it to early joint damage. In terms of scrutiny, however, the fifth MTP joint provides particular advantages, as it is superficial and readily accessible to imaging. Therefore, it is ideally suited to US examination at the patient's bedside or in the clinic.

Several studies have indicated that US is more sensitive than conventional radiography in the detection of joint damage (5, 6). Although US provides less detail compared with MRI, it nevertheless offers valuable insights into articular and periarticular pathology, thereby helping to guide treatment (13). In recent years, musculoskeletal US has become increasingly accessible to rheumatologists as a bedside diagnostic tool that can be performed as part of an outpatient assessment (14). Although US examination of all inflamed joints would be time-consuming in the clinic setting (7), selection of the joints most likely to yield pathology, such as the fifth MTP joint, could have significant diagnostic benefit and help early therapeutic decision making.

The development of joint damage is thought to result from the influence of local inflammatory cells on the cytokine and enzymatic milieu, causing cartilage destruction and bony erosion (15). Therefore, it is assumed that ongoing synovitis leads to joint erosions. Although this study noted a correlation between the presence of US-detected synovitis and radiographic erosions, US detected more cases of erosions than synovitis. It is possible that US was not sufficiently sensitive to identify synovial inflammation in every case, particularly in a small joint where erosions are more easily identifiable as breaks in the cortical surface of bone rather than soft tissue abnormalities. Alternatively, the synovitis may have preceded the development of bone damage in a condition whose time course is unknown, particularly in early disease. Treatment may also have influenced the US findings, since 11 of the 20 patients without evidence of synovitis had received corticosteroids within 4 weeks of assessment. In this regard, although 53% of patients with US-detected erosions of the fifth MPT joint had a PD signal, this finding did not reach statistical significance (P = 0.12), perhaps because erosions are a retrospective reflection of an inflammatory process, while a PD signal suggests current inflammation.

In this study, the presence of fifth MTP joint erosions did not differentiate RA from UA, but did highlight patients with otherwise undetected joint damage and future erosive potential. A larger cohort and a more prolonged study period may demonstrate such a distinction. A comparison of the total number of erosions between radiography and US was not performed since this study was undertaken primarily to demonstrate the ease with which targeted US can be utilized in a clinical setting and to identify a marker of poor prognosis at an earlier stage than can be demonstrated by radiography.

The current ACR criteria for the diagnosis of RA do not include inflammatory markers, disease activity scores such as the DAS28, anti-CCP antibody positivity, or US-detected erosions (9). However, recent evidence supports inclusion of these variables in any future diagnostic criteria (3, 16–18). We believe that this study contributes to the current knowledge in the field of both early inflammatory arthritis and the application of US in rheumatology clinics.

In patients who present with undifferentiated inflammatory arthritis, US of the fifth MTP joint identifies patients with early joint damage and has the potential to help in the diagnosis of RA where the presence of joint erosions is part of the current diagnostic criteria. This study shows that targeted US of this joint is a quick screening tool for the detection of bony damage and is more sensitive than conventional radiography in this regard. Longitudinal studies with larger patient cohorts would be useful. However, as US becomes more widely available in the rheumatology clinic setting, examination of the fifth MTP joint may yield rapid and valuable information to guide early therapeutic decisions.


  1. Top of page
  2. Abstract
  3. Introduction

All authors were involved in drafting the article or revising it critically for important intellectual content, and all authors approved the final version to be submitted for publication. Dr. Sheane had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Study conception and design. Sheane, Miller, Cunnane.

Acquisition of data. Sheane, Beddy, O'Connor, Miller.

Analysis and interpretation of data. Sheane, Beddy, Cunnane.


  1. Top of page
  2. Abstract
  3. Introduction

Abbott Laboratories had no role in the study design, data collection, data analysis, or writing of the manuscript. Publication of this article was not contingent on the approval of Abbott Laboratories.


  1. Top of page
  2. Abstract
  3. Introduction