ORIGINAL ARTICLE

Improvement in diagnosis and management of musculoskeletalconditions with one-stop clinic-based ultrasonography

Sumeet Agrawal Æ Shweta S. Bhagat ÆBhaskar Dasgupta

Received: 26 June 2008 / Accepted: 15 August 2008 / Published online: 7 October 2008

� Japan College of Rheumatology 2008

Abstract We evaluated the impact of clinic-based mus-

culoskeletal ultrasonography (MSUS) on diagnosis and

management of cases as seen in day-to-day rheumatology

practice. Data were retrieved for demography, background

condition, clinical findings, indications, regions scanned,

and outcomes of MSUS, and categorised as: new-patients

and follow-up. New-patient records were analysed as to

whether MSUS had helped to confirm or change clinical

diagnosis or was of no additional help. In follow-ups, we

determined whether MSUS had helped in disease assess-

ment, detection of co-existing problems or revision of

diagnosis. Its impact on treatment decisions was noted. A

total of 237 patients (146 women; mean age 55.9 ±

17.2 years) had 264 regions scanned; hands,50.7%. In

78/237 (32.9%) there was disagreement between clinical

and MSUS findings. Amongst new-patients (72), 13/39

(33.3%) referred with inflammatory arthritis had no MSUS

evidence of inflammation in or around joints. In 76.3% it

helped in confirming or changing diagnosis. Of the follow-

ups (165), in 78.7%, 13.9% and 7.2% it helped in assess-

ment, detection of co-existing problems and revision of

diagnosis, respectively. MSUS influenced treatment in

45/165 (27.27%) cases. In 60/67 (89.55%) cases of rheu-

matoid arthritis (RA), it was done for disease assessment;

in 31/60 (51.66%) it influenced treatment. MSUS, as a

clinic-based service in rheumatology, has significant

impact on the diagnosis and treatment of patients. This has

potential to reduce diagnostic uncertainty and follow-up

visits and ensure better outcomes.

Keywords Arthritis � Musculoskeletal � Rheumatic �Rheumatoid arthritis � Ultrasonography

Introduction

The use of musculoskeletal ultrasonography (MSUS) for

the assessment and management of musculoskeletal dis-

orders is not new. Its utility in various rheumatic diseases

has been fairly well studied. It has been shown to be at least

as sensitive as magnetic resonance imaging (MRI) in

detecting synovitis and erosions [1]. It can detect sub-

clinical inflammation in joints [2], improve joint injection

and aspiration techniques [3], and has given new insights

into the spectrum and pathogenesis of diseases [4, 5].

Rheumatologists practising MSUS often consider this as an

extension of their clinical skills. Its use for diagnosis and

intervention could possibly contribute to better and more

efficient patient care.

MSUS is becoming increasingly popular in rheumatol-

ogy clinics, and the number of research publications related

to its use by rheumatologists has increased. Most of these

studies are, however, from specialized rheumatology units

and have focused on specific diseases or particular aspects

of them.

Despite much being done, its utility in routine rheuma-

tology practice needs to be further justified, especially with

regards to its impact on patient treatment. We undertook

this study in a large district general hospital dealing with a

varied mix of inflammatory and non-inflammatory rheu-

matic diseases with an aim to look at the impact of MSUS

S. Agrawal

Department of Rheumatology, Nizam’s Institute of Medical

Sciences, Hyderabad, India

S. S. Bhagat � B. Dasgupta (&)

Department of Rheumatology, Southend University Hospital,

NHS Trust, Prittlewell Chase, Westcliff-on-Sea,

Essex SS0 0RY, UK

e-mail: Bhaskar.Dasgupta@southend.nhs.uk

123

Mod Rheumatol (2009) 19:53–56

DOI 10.1007/s10165-008-0122-4

on patient management in a real-life clinic setting as done

by rheumatologists as a one-stop procedure.

Patients and methods

Data from all patients who had undergone MSUS in gen-

eral rheumatology clinics, Department of Rheumatology,

Southend University Hospital NHS Trust, between January

and December 2007 were retrieved. The scans were done

by consultant rheumatologists trained and experienced in

MSUS and were done during the same visit. Patient

selection for scanning was at the discretion of the clini-

cians. Scans were performed on an Esaote-Mylab70

machine (Esaote SpA, Genoa, Italy) using a linear trans-

ducer (7–15 MHz). Demographic details, background

condition, clinical findings, regions scanned and indica-

tions and outcomes of MSUS assessment were recorded

from patients’ case notes and clinic letters. For the purpose

of defining the regions scanned, ‘hands’ included the hands

and wrists, ‘ankle’ included ankle and heel and ‘feet’

included midfoot and forefoot. Shoulder, elbow, hip and

knees implied the respective anatomical regions.

The indications of MSUS are different in new patients

(undiagnosed rheumatic disorder) compared with those

with known conditions coming for follow-up. Patients were

analysed in two groups: group 1 comprised new patients

without definite diagnosis of a rheumatic disease; group 2

was composed of follow-up patients with a definite diag-

nosis. On the basis of the impact MSUS had in each group,

they were further categorised. Group 1 categories: A-

MSUS, helped in confirming clinical diagnosis; B-MSUS,

helped in changing the clinical diagnosis; C-MSUS, of no

additional help over clinical evaluation. Group 2 catego-

ries: A-MSUS, helped in disease assessment; B-MSUS,

helped in detecting co-existing musculoskeletal problems;

C, revision of existing diagnosis based on MSUS. In

group 2, data were also collected about the way MSUS

influenced treatment decisions. Classification into these

categories was arrived at after mutual agreement between

two rheumatologists.

Results

There were 237 patients studied (91 men, 146 women);

their mean age was 55.9 ± 17.2 years (range 9–97 years).

A total of 264 regions were scanned: hands, 134 (50.7%);

shoulder, 44 (16.6%); ankle, 32 (12.8%); knee,19 (7.1%);

feet,15 (5.6%); others (elbow, 8; hip, 6; vascular, 3; sterno-

clavicular joint, 1; ischium, 1; breast, 1). In 25 patients

more than one region was scanned (not including the

contralateral region). Overall, for 78 patients (32.9%),

there was disagreement between clinical findings and

MSUS. Of these, 111/134 (82.8%) of the hands scanned

were related to inflammatory aetiology. Hands were part of

all scans for polyarthritis and arthralgias or otherwise

wherever indicated.

Group 1 constituted 72 of a total of 800 (9%) new

patients seen during the period (Table 1). Of these, 39

(54.2%) had been referred for inflammatory arthritis (pol-

yarthritis, 23; oligoarthritis, 7; monoarthritis, 9). Others

included arthralgias, 10 (13.9%); shoulder pains, 13

(18.0%); tendinopathy, 5 (6.9%); carpal tunnel syndrome, 3

(4.2%); cellulitis and bursitis, 1 (1.4%) each. In 13/39

(33.3%) patients clinically suspected to have inflammatory

arthritis, MSUS did not reveal any evidence of inflamma-

tion (effusion, synovial hypertrophy or Doppler signals) in

the joints or juxta-articular structures. Data on inflamma-

tory markers and rheumatoid arthritis (RA)-associated

antibodies were retrieved for these 13 patients. C-reactive

protein (CRP) level was raised in two patients (55 mg/l and

23 mg/l, respectively; normal 0–10 mg/l). Findings for

immunoglobulin M–rheumatoid factor (IgM-RF) were

positive in low-titres in two patients (19 IU/ml and 15 IU/

ml; negative finding \14 IU/ml; both these patients were

different from the ones with raised CRP levels). The test

results for Westergren’s erythrocyte sedimentation rate

(ESR) was available for six patients, and the rate was

raised in one (43 mm in the 1st hour; normal is 0–37 mm

in the 1st hour; this was the same patient with a CRP level

of 55 mg/l). An anti-cyclic citrullinated peptide (anti-CCP)

antibody assay was done for only one patient (with both

CRP and ESR raised) and the result was negative (1 U/ml;

normal 0–10 U/ml). In none of the patients referred for

arthralgias could MSUS demonstrate inflammation. In all

patients with shoulder pains at least one abnormality

(rotator-cuff disorder, osteoarthritis and/or bursitis) could

be detected. The influence of MSUS on group 1 patients is

summarized in Table 1. Overall, in 76.3% of new patients

(groups 1A ? 1B), MSUS had helped the rheumatologist

to arrive at a diagnosis.

There were 165 patients in group 2 (Table 1). Rheu-

matoid arthritis, osteoarthritis, rotator-cuff disorders,

spondyloarthropathies, polymyalgia rheumatica and juve-

nile idiopathic arthritis accounted for the majority, with 67

(46.6%), 16 (9.7%), 13 (7.9%), 11 (6.6%), 10 (6.0%) and 9

(5.4%) patients, respectively. The impact of MSUS on

group 2 patients is summarized in Table 1. MSUS findings

influenced treatment decisions in 45/165 (27.3%) of the

patients (Table 2).

In RA specifically, 60/67 (89.5%) patients underwent

scanning for disease assessment, and in 31/60 (51.7%)

patients treatment was influenced on the basis of MSUS

findings. These patients constituted the majority of the 45

patients in whom MSUS findings had influenced treatment.

54 Mod Rheumatol (2009) 19:53–56

123

Discussion

This is pragmatic data from a rheumatology clinic, evalu-

ating the real-life role of MSUS. It shows the utility of

MSUS for a variety of disorders and across a wide age

range. The most frequently scanned region was the hand

(including the wrist). In a substantial number there was

disagreement between clinical and MSUS findings. In the

majority of new patients MSUS helped the rheumatologist

to arrive at a diagnosis. However, in approximately one-

third of patients referred for inflammatory arthritis, it did

not reveal any evidence of inflammation in or around the

joint. Of the follow-up patients, in 80% it was used for

assessment of disease, especially in RA. MSUS also

influenced treatment in considerable numbers. However,

time and resource constraints meant that only a small

proportion of patients seen in the clinic could actually

undergo the procedure.

The benefits of MSUS in the assessment of musculo-

skeletal conditions are well known [6–8]. Karim et al. have

reported that following ultrasonography the site-specific

diagnosis had been changed in 53% of their patients [9].

This is also evident by the more than 30% disagreement we

found between clinical and MSUS findings.

In inflammatory arthritis, the best practice currently

emphasises early evaluation by rheumatologists, especially

since discrepancy between referral diagnoses and diagno-

ses made by rheumatologists is well known [10, 11]. In

33.3% of new patients in our study with clinical suggestion

of inflammatory arthritis, MSUS did not reveal any evi-

dence of inflammation in or around the joints. This can

have considerable impact on subsequent evaluations (with

its economic implications), treatment considerations and

counselling of patients.

In a significant number of new patients, MSUS helped

the rheumatologist to arrive at a diagnosis, suggesting that

it improves diagnostic confidence [9, 12]. This may mean

decreased reliance on other imaging modalities and

reduced delay in instituting appropriate therapy.

Amongst follow-up patients, MSUS was mainly used to

assess the disease or to detect co-existing musculoskeletal

problems. In the majority, MSUS findings influenced the

treatment and altered the management plan, as was also

noted by Karim et al. In 7% of follow-up patients it also led

to revision of the existing diagnosis. Comparable values

have been reported in another study [9].

Because of the prospects of disease modification in RA,

proper assessment of disease activity is important. This is

probably the reason why 90% of RA patients who under-

went MSUS were offered the investigation for assessment

of disease activity. Interestingly, more than half of these

MSUS findings influenced treatment. Karim et al. have also

reported that the management plan was altered in 53% of

their patients based upon MSUS findings [9]. Together,

both these studies stress that MSUS has considerable

impact on management decisions, and this may have likely

influence on long-term outcomes. Larger prospective

studies could clarify this issue further.

To conclude, our study highlights the significant positive

impact of integrating MSUS as a one-stop service in rou-

tine rheumatology clinic. We feel it has the potential for

increasing effectiveness of care by supplementing clinical

evaluation, limiting diagnostic uncertainty and reducing

reliance on other imaging services.

The study has the limitation of being a retrospective

analysis. Because of the next-door availability of the

Table 1 The sites scanned and the impact of MSUS in the different

groups

Parameter Group 1 Group 2

Total no. of patients (female:male) 72 (50:22) 165 (97:68)

Mean age ± standard deviation (SD)

(years)

54.08±15.43 56.69±17.90

Total no. of regions scanned 81 183

Hands 39 (48.1%) 95 (51.9%)

Shoulder 16 (19.7%) 28 (15.3%)

Ankle 14 (17.3%) 18 (9.8%)

Others 12 (14.8%) 42 (22.9%)

Categories

A 47 (65.2%) 130 (78.7%)

B 8 (11.1%) 23 (13.9%)

C 17 (23.6%) 12 (7.2%)

Group 1: A-MSUS helped in confirming clinical diagnosis, B-MSUShelped in changing the clinical diagnosis, C-MSUS of no additional

help over clinical evaluation

Group 2: A-MSUS helped in disease assessment, B-MSUS helped in

detecting co-existing musculoskeletal problems, C revision of exist-

ing diagnosis based on MSUS

Table 2 Influence of MSUS on treatment in group 2 patients

Treatment decision Number of patients

DMARD alterationa 26 (57.7%)b

DMARD startedc 5 (11.11%)

Systemic corticosteroidd 15 (33.33%)

Local corticosteroide 14 (31.11%)

a Change in the dose/route of existing disease-modifying anti-rheu-

matic drug (DMARD) and/or add-on DMARD (RA patients)b In three patients it led to the consideration of anti-tumour necrosis

factor (anti-TNF) therapyc DMARD started for the first time based on MSUS report (RA

patients)d Orally or intra-muscularly (RA patients)e Intralesional/intra-articular administration

Mod Rheumatol (2009) 19:53–56 55

123

machine, the bias in selecting patients for scanning cannot

be ruled-out. The economic implications and the impact of

introducing MSUS on staff and resources and day-to-day

running of the clinic have not been looked into. Further

work needs to be done so that we can understand how this

would translate into long-term outcomes and gains.

Acknowledgments Sumeet Agrawal was on an Asia Pacific League

of Associations for Rheumatology (APLAR) 2007 fellowship at the

Department of Rheumatology, Southend University Hospital NHS

Trust.

Conflict of interest statement The authors declare that they have

no conflict of interest.

References

1. Szkudlarek M, Klarlund M, Narvestad E, Court-Payen M,

Strandberg C, Jensen KE, et al. Ultrasonography of the meta-

carpophalangeal and proximal interphalangeal joints in

rheumatoid arthritis: a comparison with magnetic resonance

imaging, conventional radiography and clinical examination.

Arthritis Res Ther. 2006;8:R52.

2. Wakefield RJ, Green MJ, Marzo-Ortega H, Conaghan PG,

Gibbon WW, McGonagle D, et al. Should oligoarthritis be

reclassified? Ultrasound reveals a high prevalence of subclinical

disease. Ann Rheum Dis. 2004;63:382–5.

3. Galiano K, Obwegeser AA, Walch C, Schatzer R, Ploner F,

Gruber H. Ultrasound-guided versus computed tomography-

controlled facet joint injections in the lumbar spine: a prospective

randomized clinical trial. Reg Anesth Pain Med. 2007;32:317–22.

4. Ceccato F, Roverano SG, Papasidero S, Barrionuevo A, Rillo OL,

Paira SO. Peripheral musculoskeletal manifestations in polym-

yalgia rheumatica. J Clin Rheumatol. 2006;12:167–71.

5. Frediani B, Falsetti P, Storri L, Bisogno S, Baldi F, Campanella

V, et al. Evidence for synovitis in active polymyalgia rheumatica:

sonographic study in a large series of patients. J Rheumatol.

2002;29:123–30.

6. Rees JD, Pilcher J, Heron C, Kiely PD. A comparison of clinical

vs ultrasound determined synovitis in rheumatoid arthritis uti-

lizing gray-scale, power Doppler and the intravenous

microbubble contrast agent ‘Sono-Vue’. Rheumatology. 2007;46:

454–9.

7. Gisondi P, Tinazzi I, El-Dalati G, Gallo M, Biasi D, Barbara LM,

et al. Lower limb enthesopathy in patients with psoriasis without

clinical signs of arthropathy: a hospital-based case-control study.

Ann Rheum Dis. 2008;67:26–30.

8. Kane D, Grassi W, Sturrock R, Balint PV. Musculoskeletal

ultrasound—a state of the art review in rheumatology. Part 2:

Clinical indications for musculoskeletal ultrasound in rheuma-

tology. Rheumatology. 2004;43:829–38.

9. Karim Z, Wakefield RJ, Conaghan OG, Lawson CA, Goh E,

Quinn MA, et al. The impact of ultrasonography on diagnosis and

management of patients with musculoskeletal conditions.

Arthritis Rheum. 2001;44:2932–5.

10. Bolumar F, Ruiz MT, Hernandez I, Pascual E. Reliability of the

diagnosis of rheumatic conditions at the primary health care level.

J Rheumatol. 1994;21:2344–8.

11. Courtney PA, Wright GD. Referrals to an ‘early synovitis clinic’:

are they appropriate? Ann Rheum Dis. 2001;60:991–2.

12. Grassi W. Clinical evaluation versus ultrasonography: who is the

winner? J Rheumatol. 2003;30:908–9.

56 Mod Rheumatol (2009) 19:53–56

123