British Journal of Rheumatology 1996;35:676-679

ULTRASONOGRAPHIC FEATURES OF DIABETICCHEIROARTHROPATHY

A. A. ISMAIL, B. DASGUPTA, A. B. TANQUERAY and J. J. HAMBLINSouthend Hospital, Westcliff-on-Sea, Essex

SUMMARYUltrasonography was used to measure flexor tendon sheath thickness in 14 insulin-dependent (IDDM) diabetics with diabeticcheiroarthropathy (DCA) and compared to 17 IDDM patients without DCA along with 10 healthy volunteers. Assessment wasalso made of the presence of systemic diabetic microvascular disease complications. A blinded visual 'eyeball' report on theultrasound scans by a radiologist found hypoechoic thickening of the flexor tendon sheaths in 12 of the 14 patients with DCA,three of the 17 unaffected diabetics and two of the healthy volunteers (Fisher's exact, P < 0.001). However, further quantitationof tendon sheath thickness separated patients with DCA from others. In all patients with DCA, tendon sheath thickness was> 1 mm (median 1.8 mm, range 1.0-2.3 mm) and < 1 mm in the other two groups (medians 0.6 and 0.5 mm, range 0.3-1.0 mm)(Kruskal-Wallis, P < 0.001). All patients with DCA had evidence of systemic microvascular disease complications, particularlyproliferative retinopathy (82%). It appears that flexor tendon sheath thickening in the hand is an integral part of the pathologyin DCA and is easily demonstrated by ultrasound. It is closely associated with overt diabetic raicrovascular disease complications.

KEY WORDS: Diabetic cheiroarthropathy, Tendon sheath, Insulin-dependent diabetes, Diabetic microvascular complications.

A LARGE proportion of morbidity and mortality due todiabetes mellitus is due to chronic microvascularcomplications such as retinopathy, nephropathy andneuropathy. An under-recognized complication isdiabetic cheiroarthropathy (DCA). DCA is a syndromecharacterized by painless limitation of the finger jointsand thick, tight, waxy skin. The cardinal feature is the'prayer sign'. Studies largely agree on a prevalencefigure of 30-35% and it may be related to othermicrovascular disease complications [1].

The aims of this study were to assess (1) the flexortendon and tendon sheath using ultrasound and (2) thepresence of systemic diabetic microvascular diseasecomplications in patients with DCA.

SUBJECTS AND METHODSThe diagnosis of DCA was made independently by

two clinicians using the 'prayer sign', characterized byincomplete approximation of one or more of the digitswhen the patient attempts apposition of the palmarsurfaces of the proximal and distal interphalangealjoints with palms pressed together and the fingersfanned. Particular attention was also given to the skinof the hands.

Fourteen insulin-dependent diabetes mellitus(IDDM) patients with DCA were identified from thediabetic clinic at Southend Hospital (male = 8,female = 6, mean age 43.4 yr, age range 27-54 yr). Agroup of 17 insulin-dependent diabetics (male = 9,female = 8, mean age 41.2 yr, age range 27-59 yr) withno evidence of DCA were included in the study, alongwith 10 healthy volunteers with no evidence of diabetesmellitus. The diabetics without DCA were matched asclosely as possible for age ( ± 5 yr), sex and duration

Submitted 25 July 1995; revised version accepted 2 February 1996.Correspondence to: B. Dasgupta, Southend Hospital, Prittlewell

Chase, Westdiff-on-Sea, Essex, SS0 0RY.

of diabetes (DCA group: mean duration = 26 yrand range 10-40 yr; non-DCA group: meanduration = 25.4 yr and range 10-43 yr). Details of thepatients' insulin dose and diabetic control, as measuredby fructosamine levels in the preceding 2 yr, wereobtained from the diabetic clinic records.

High-frequency ultrasonography using a 7.5 MHz,and subsequently a 10 MHz, probe was used toexamine the volar aspects of both hands. Theexamination was performed in the transverse planewith special attention paid to the evaluation of allflexor tendons and tendon sheaths in the hand, but onlythe tendon sheaths with the greatest thickness wereused for measurements used in the analysis. First, allthe ultrasound scans underwent a simple blinded visual'eyeball' report by a consultant radiologist (ABT) andthe measurements of the tendon sheaths performedusing the facilities on the ultrasound scanner.

Retinopathy was assessed by fundoscopy andscreening retinal photography performed in thediabetic clinic. Peripheral neuropathy was assessed byclinical examination of the ankle jerks and vibrationsense, and nephropathy by the presence of persistentproteinuria and/or impaired renal function(creatinine > 125/imol/l). Clinical examination wasperformed for Dupuytren's contracture and anyunderlying joint disease, and these subjects were notincluded in the study if present. In addition, flexortenosynovitis (palpable crepitus) and symptoms andsigns of carpal tunnel syndrome (Tinel's sign) weresought.

RESULTSUltrasound scans from all patients with DCA

showed a hypoechoic thickening of the flexor tendonsheaths. Quantitative analysis of the tendon sheaththickness from DCA patients revealed measure-ments > 1 mm (median 1.8 mm, range 1.0-2.3 mm)

© 1996 British Society for Rheumatology676

ISMAIL ET AL.: ULTRASOUND OF THE DIABETIC HAND 677

(Fig. 1), compared to diabetic patients without DCAand healthy controls (< lmm, medians 0.6 and0.5 mm, respectively, range 0.3-1 mm) (Fig. 2).Figure 3 shows the medians and 95% confidenceintervals of tendon sheath measurements in the threegroups (Kruskal-Wallis test, P < 0.001).

The ultrasound scans also underwent blinded visualreport by the radiologist and tendon sheath thickening

was found in 12 of the 14 patients with DCA, and inonly three of the unaffected diabetics and two of thehealthy volunteers {P < 0.001, Fisher's exact test). Nodifference in the tendon itself was observed between thegroups. Specifically, there was no evidence of tendonrupture or any other soft-tissue abnormalities.

All the 14 patients with DCA had systemicmicrovascular disease, particularly proliferative

PHR: 8SX

MEASUREB-l

FIG. 1-

3LN 1. SMPWR: 80S

MEASUREB-l

FlO. 2.

Fio. 1 and 2. Ultrasound scans of a patient with diabetic cheiroarthropathy (Fig. 1) and an unaffected diabetic control (Fig. 2) showinghypoechoic thickening of the flexor tendon sheath (shown by the two crosses) in diabetic cheiroarthropathy (thickness 2.3 mm) compared tothe control (thickness 0.6 mm)

678 BRITISH JOURNAL OF RHEUMATOLOGY VOL. 35 NO. 7

2.1

1.8

1.5

1.2

0.9

0.6

0.3

• M B

m

-

DCA

• •• •

• • • - - -•NON-UM NUN-DuA

Fio. 3.—Graphical representation of measurements of tendon sheath thickness in the DCA group, healthy volunteers (non-DM) and diabeticcontrols (non-DCA). The median and 95% CI are shown.

retinopathy in 13 out of 14 patients (82%), neuropathyin five patients and ncphropathy in five patients(Table I). On the other hand, in the IDDM groupwithout DCA, four had evidence of microvasculardisease and only one patient had proliferativeretinopathy (Fisher's exact test, P < 0.001) (Table I).No significant differences were observed between

TABLE IComplications and associated clinical features in patients with and

without diabetic cheiroarthropathy (DCA)

Microvascularcomplications*

Proliferativeretinopathy

NeuropathyNephropathy

Diabetic controlfFnictosamine(median, 95% CI)

Associated clinicalfeatures*

Carpal tunnelsyndrome

Flexortenosynovitis

IDDMwith DCA

n - 14

13/145/144/14

355(CI 347-378)

4/14

2/14

IDDMwithout DCA

71=. 17

1/171/172/17

377( O 330-388)

3/17

0/17

Significancelevel

P < 0.001P - 0.05/> = 0.23

P = 0.9

P- 0.383

? = 0.19

•Fisher's exact testtMann-Whitney test.

groups for carpal tunnel syndrome, flexor teno-synovitis, and diabetic control or insulin requirementsover the preceding 2 yr (Table I).

DISCUSSIONUsing ultrasound, the thickness of the flexor tendon

sheath appears to be significantly more profound indiabetics with DCA when compared with IDDMpatients without DCA and non-diabetic controls. Thissuggests that flexor tendon sheath thickening is anintegral part of DCA. The pathogenesis of DCA is stillnot known, but increased skin thickness has beendemonstrated both histologically [2] and usingultrasound [3]. Glycosylation of collagen is thought tobe important in the pathogenesis of increased skinthickness in DCA [4].

Real-time high-frequency ultrasound has not beenused to evaluate the tendons and tendon sheath inDCA. It proved to be an easy and quick investigationto perform on the diabetic hand, and effectivelydemonstrated tendon sheath thickening, and theresolution was significantly improved using the10 MHz probe.

A significant association was found between DCAand systemic microvascular complications, particularlyproliferative retinopathy. The difference in theprevalence of microvascular disease complications inthe two groups may be attributed to genetic differences

ISMAIL ET AL.: ULTRASOUND OF THE DIABETIC HAND 679

[5]. It is also important to point out that diabeticcontrol was documented in our patients only in the 2 yrprior to the study. The degree of diabetic control beforethis period could not be established and may in facthave influenced the disease progress in the DCA group.However, it appears abundantly clear that DCA sharesa common pathogenetic link with systemicmicrovascular complications. Further work is neededto determine whether hand ultrasound could providean easy, early test for microvascular disease which maycomplement conventional screening methods, e.g.retinal screening.

In conclusion, flexor tendon sheath thickeningappears to be a major component of DCA, and is easilydemonstratable on ultrasound. DCA is closelyassociated with microvascular disease.

REFERENCES

1. Rosenbloom AL, Silverstein JH, Lezotte DC,Richardson K, McCallura M. Limited joint mobility inchildhood diabetes mellitus indicates increased risk formicrovascular disease. N Engl J Med 1981;305:191-4.

2. Rosenbloom AL. Skeletal and joint manifestations ofchildhood diabetes. Pediatr Clin North Am 1984;31:569-89.

3. Collier A, Matthews DM, Kellett HA, Oarke BF,Hunter JA. Change in skin thickness associated withcheiroarthropathy in insulin dependent diabetes mellitus.BT Med J 1986;292:936.

4. Isdale AH. The ABC of the diabetic hand—advancedglycosylation end products, browning and collagen[Editorial]. BT J Rheumatol 1993^32:859-61.

5. Pyke DA, Tatersall RB. Diabetic retinopadiy in identicaltwins. Diabetes 1973;22:613-8.