Computed tomography in the assessment of response to limb compression in unilateral lymphoedema

Computed tomography in the assessment of response to limb compression in unilateral lymphoedema

Clinical Radiology (1995) 50, 541-544 Computed Tomography in the Assessment of Response to Limb Compression in Unilateral Lymphoedema C. D. COLLINS, ...

944KB Sizes 0 Downloads 30 Views

Clinical Radiology (1995) 50, 541-544

Computed Tomography in the Assessment of Response to Limb Compression in Unilateral Lymphoedema C. D. COLLINS, P. S. MORTIMER*, H. D'ETTORRE, R. P. A'HERNi" and E. C. MOSKOVIC

Departments of Radiology, *Dermatology and ~Computing and Information, Royal Marsden Hospital, London, UK In this prospective study computed tomography (CT) was used to monitor the response of compression therapy in 27 patients with chronic unilateral lymphoedema over a 12 week period. Computed tomography examination of abnormal and normal limbs (proximal and distal portions) was performed in the first, third and 12th weeks of treatment. Changes in cross-sectional area (CSA) and average densities of the different compartments within the proximal and distal portions of the abnormal limb were compared with the normal side. The most significant decrease in CSA occurred within the subcutaneous compartment of the distal portion (P= 0.002); the decrease in CSA of the proximal portion was also significant (P= 0.02) but changes in muscle and bone compartments were not significant. Significant differences in average density measurements of the subcutaneous and muscle compartments remained between normal and abnormal limbs following the conclusion of the study (P = 0.001 and P = 0.01, respectively). This study demonstrates that CT is a useful method for monitoring therapeutic response to compression therapy. Collins, C.D., Mortimer, P.S., d'Ettorre, H., A'Hern, R.P. & Moskovic, E.C. (1995). Clinical Radiology 50, 541-544. Computed Tomography in the Assessment of Response to Limb Compression in Unilateral Lymphoedema

Acceptedfor Publication 4 April 1995

Lymphoedema is a well-recognized cause of chronic limb swelling and arises when there is failure of lymph drainage [1,2]. Clinically it is subdivided into primary (congenital) and secondary forms [3,4]. Secondary lymphoedema is more common and occurs following damage or obstruction to the lymphatic system as a result of surgery/trauma, irradiation or, in tropical areas, chronic filarial infestation. Venous abnormalities often coexist and also contribute to swelling [5]. Ann swelling following surgery and/or radiotherapy for breast cancer occurs in 2% to 60% of patients (depending on the operative procedure) [6,7] and may be very disabling. It is the essential function of the lymphatic system to drain the tissues of macromolecules, particularly proteins which are too large to re-enter the blood stream directly [8]. Therefore, in lymphoedema, unlike all other forms of non-inflammatory oedema, it is not just water which contributes to swelling. Clinically this manifests as a brawny or more solid form of oedema which is more difficult to treat. Computed tomography (CT) of lymphoedematous limbs has demonstrated a characteristic 'honeycomb' pattern in the subcutaneous compartment [4,9,10]. A previous study quantified the changes in crosssectional area (CSA) of subcutaneous tissues following surgery or massage therapy in the distal portion of affected limbs [11]. However, no attempt has been made to quantify the changes in either the CSA or CT densities of the various tissue compartments. The aim of the current study was to assess the efficacy of CT in analysing these changes following different forms of compression therapy. Correspondence to: Dr C. D. Collins, Department of Radiology, Christie Hospital NHS Trust, Wilmslow Road, Manchester M20 4BX, UK. 9 1995 BlackwellScienceLimited

PATIENTS AND METHODS Approval to undertake this study was granted by the Hospital Ethics Committee. Twenty-seven patients with unilateral limb oedema were recruited prospectively. The study included six patients with primary and 21 patients with secondary unilateral lymphoedema. The latter group were free from the original disease at the time of the study and had not undergone any form of treatment (surgery, radiotherapy or chemotherapy) for at least 12 months. The average duration of limb swelling was 6.7 years. For inclusion into the study the abnormal limb had to be at least 20% larger (as measured by plethysmography [12]) than the normal (control) side. All patients received compression therapy during the period of the study. Treatment comprised either 3 weeks of multi-layer bandaging initially followed by elastic hosiery or 12 weeks of elastic hosiery only. Computed Tomography Scanning Protocol Patients were examined on day 1, day 21 and after 12 weeks using a Somatom DR2 (Siemens, Erlangen) scanner. Axial slices (8 mm) were obtained at predetermined fixed points at two equivalent levels in both the abnormal and normal limb. Exact levels for scanning each limb were obtained by measuring a specified distance from a fixed anatomical point (radial styloid, ulnar olecranon, medial malleolus) on each occasion so that scans could be directly compared following each examination. The circumferences of the whole limb, muscle mass and bones were traced using a digitizer pen linked to the CT computer. The difference between the whole limb and the muscle mass provides an absolute measurement of the CSA of the subcutaneous compart-

542

CLINICAL RADIOLOGY

ment. This area was calculated before and after treatment and the decrease in CSA was expressed as a percentage of the initial area for each limb [11]. The mean CT (Hounsfield Unit) densities of the subcutaneous and muscle compartments on both sides were similarly calculated [10]. Statistical Methods Data analysis was performed using MINITAB | [13] software. The CSA and density values were standardized by dividing by the mean of the control values. Nonparametric statistical tests were used and median values are quoted. The Wilcoxon signed-rank test was used to examine whether changes in CSA and density in the abnormal limb were significant when compared with the normal side.

portion and 36% greater in the proximal portion (P < 0.001 for both differences) when compared with the normal limb. This result was due to the CSA of the subcutaneous compartment remaining 157% and 74% greater for the distal and proximal portions, respectively than the normal side (both P < 0.001). No statistically significant change was seen in the muscle or bone compartments between the two limbs. Both the proximal and distal portions of the abnormal limb demonstrated a significant decrease in overall standardized CSA over the time period of the study (median decrease: distal portion 7%, P = 0.005; proximal

RESULTS The study group of 27 patients comprised 23 females and four males (range 24-86 years, mean 56 years). Twenty-one patients had lymphoedema secondary to a variety of causes of which previous surgery and radiation to the breast and axilla was the most common (n = 9) (Table 1). In six patients the lymphoedema was primary (congenital). The abnormal limb was the arm in 18 patients and the leg in nine patients. All patients demonstrated characteristic CT imaging features of lymphoedema in the the abnormal limb, i.e. increased skin thickness and soft tissue stranding in the form of a 'honeycomb' pattern in the subcutaneous compartment (Fig. 1). Fluid bound by outer muscle fascial layers was present in 23/27 distal and in 4/27 proximal portions of abnormal limbs.

(a)

Reproducibility of Technique The regions of interest on both the abnormal and normal limbs (proximal and distal portions) were outlined by the same person in all instances in order to keep any operator variation to a minimum. Analysis of the results from the normal limb demonstrated little variation in CSA or density readings over the course of the study. Therefore any change in CSA or density on the abnormal side can be deemed to have arisen as a result of treatment and not as a result of operator variation.

(b)

Standardized Cross-Sectional Area At 12 weeks, the overall standardized CSA of the abnormal limb was still 38% greater in the distal Table 1 - Aetiologieal factors present in patients with secondary lymphoedema

Aetiological factors Surgery and radiation to breast and axilla Surgery to breast alone, no radiation Surgery to breast alone, radiation to axilla Surgery and radiation to breast alone Surgery to breast and axilla, no radiation Radiation only (pelvic) Total

No. 9 4 4 2 1 1 21

(c) Fig. 1 - (a-c) Computed tomography sections of distal upper limb at 1, 3 and 12 weeks in a patient undergoing compression therapy (3 weeks of multi-layer bandaging followed by elastic hosiery). The decrease in amount of fluid surrounding muscles (arrows) and in degree of skin thickness is already apparent following the second examination. 9 1995 BlackweUScienceLtd, ClinicalRadiology, 50, 541-544.

543

ASSESSMENT OF COMPRESSION THERAPY IN UNILATERAL LYMPHOEDEMA USING CT

e, 1.75 E 1.5

tO

o 0.9 E

"10

1.25

0.es

~o

0

"

1

.

.

.

Distal muscle

.......

0.95

E

m

.

:. . . . . . .

E tO .. JO

Treated = control

1 .Q

I

0

I 3

Treated = control

I 12

Weeks after start of treatment

0

[

Proximal subcuta neo-~-~"t

......"[ ...............................~'i'st~l'i'S~b;u't; neo u s

I

0.8

er

0.75

I

0

12 Weeks after start of treatment

Fig. 2 - Change in total cross-sectional area ratio between the abnormal (lymphoedematous) and control (normal) limbs (median values + s.E. of the median). - - , Distal; - - - , proximal.

portion 2%, P = 0.02) (Fig. 2). The mean decrease in the subcutaneous compartment alone was 26% for the distal portion (P = 0.002) and 9% for the proximal portion (P = 0.02) (Fig. 3). No significant decrease was observed in either the muscle or bone compartments. Change in CT Density Measurements

At 12 weeks significant differences remained in the average density measurements between the normal and abnormal limbs. This difference was most significant for the subcutaneous compartment distally (11~ decrease, P < 0.001) and proximally (16% decrease, P = 0.001); it was also significant for the muscle compartment proximally and distally (both 2% decrease, P = 0.01) (Fig. 4). Over the time period of the study the change in density of the lymphoedematous limb was most significant in the subcutaneous compartment of the distal portion (8% decrease, P = 0.01). The change in density of the proximal subcutaneous compartment was not significant. The muscle compartment of the proximal and distal portions also showed small changes in density (P = 0.03 and P = ns, respectively). The change in density and standardized CSA correlated significantly in the distal (r = 0.62, P = 0.002) but not in the proximal (r = 0.28, P = ns) portion. There was no significant difference in the decrease of CSA and density when patients with arm lymphoedema were compared with those with leg lymphoedema.

"Q ._E

3

2 9~

2.5

Distal

J

1

O o J~

E

2 .........................

"10

$

P_roxim__a/_ __]

~.5

0 IIC

I

I

I

0

3

12

Weeks after start of treatment

Fig. 3 - Change in cross-sectional area ratio o f the subcutaneous compartment between the abnormal (lymphoedematous) and control (normal) limbs (median values + s.E. of the median). - - , Distal; - - - , proximal. 9 1995 Blackwell Science Ltd, Clinical Radiology, 50, 541-544.

Fig. 4 - Change in density measurement ratios of muscle and subcutaneous compartments between the abnormal (lymphoedematous) and control (normal) limbs (median values + s.E o f the median). , Distal muscle; - - - , proximal muscle; - - - , distal subcutaneous; . . . . , proximal subcutaneous.

DISCUSSION One of the problems in treating lyrnphoedema is a lack of sensitive and reproducible methods for evaluating therapy interventions. At present the only widely used method for monitoring response to treatment is measuring limb volume by either water displacement or multiple circumferential tape measurements. However, differences in methodology design and a relatively high error between independent measures make comparison of results between lymphoedema therapy centres meaningless. Other methods described for monitoring treatment response are tonometry [14], lymphography and lymphoscintigraphy but they are not widely available and subject to difficulties in interpretation. In this study CT has been shown to be an objective and reproducible method for evaluating changes in lymphoedema following treatment. Computed tomography not only provides information through CSA of volume change to a limb but will identify the compartment in which that change takes place. Following compression therapy a significant decrease in CSA for the limb overall occurs but more particularly for the subcutaneous compartment where most of the swelling was clearly shown to reside. The density measurements indicate change in relation to tissue characteristics within the separate compartments. The presence of fibrous tissue and free fluid within the lymphoedema probably influences average density measurements. Although no significant difference in CSA between lymphoedematous and nonswollen limbs was shown for the muscle compartments, differences in density were demonstrated possibly as a result of water molecules being bound within the structure of larger macromolecules. The ease with which a compartment can be delineated by a region of interest conveys highly reproducible results. This fact, combined with the consistency and objective nature of the methodology, makes it ideal for use when comparing results of treatment between centres such as in the case of a multicentre trial. One previous study has used CT to monitor response to therapy [11]. That study involved taking comparative measurements in the mid-calf in 20 patients with primary lymphoedema undergoing treatment with either surgery or lymphopress massage. To the best of our knowledge

544

CLINICAL RADIOLOGY

no paper has been published in the radiological literature on the value of using CT to monitor response to therapy. :The current study has measured both the proximal and distal portions of normal and abnormal limbs using a different timescale and different forms of compression therapy. The non-invasive nature of CT is a major factor in its high patient acceptability. The facility to visualize two limbs on one image and the ability to permanently store the image makes direct comparison easy. Although this technique employs radiation, measurements from this institution indicate that the absorbed dose for this CT scanning protocol is very low. The red bone marrow receives the most significant dose, and the equivalent dose for this organ is estimated to be 0.17 mSv per four 10mm slice procedure. The aggregated radiation detriment associated with this dose for three procedures is 4.2 • 10-6 assuming risk factors from ICRP 60 averaged over an adult working population [15]. Computed tomography has been used in other conditions for monitoring treatment response. Vaughan [10] described a significant change in the CSA of muscle compartments in two out of seven patients with lymphoedema following femoral-popliteal bypass surgery. Computed tomography is also regularly used to monitor changes in tumour size following chemotherapy/ radiotherapy and in assessing pulmonary parenchyma following steroid treatment in interstititial lung disease [161. It has been argued elsewhere that MRI might be more suitable because of its ability to detect water and the lack of radiation [17]. However, a pilot study with eight patients using the technique in this institution highlighted three disadvantages which preclude its use on a regular basis. The most notable was claustrophobia which affected all patients; the larger than normal body mass makes these patients particularly susceptible to this problem. The other disadvantages were difficulty in arm positioning and the length of time required to perform the examination (a function of time required for correct positioning, relief of claustrophobia and running of appropriate sequences), taking on average 60 min. In conclusion, this study demonstrates the efficacy of CT in monitoring the effects of compression therapy

within the different compartments of the lymphoedematous limb. Acknowledgements. The authors wish to thank Dr David Dance, Department of Physics, Royal Marsden Hospital, for help and advice on CT measurements and the radiographers in CT for their assistance.

REFERENCES 1 F61di E, F61di M, Clodius L. The lymphoedema chaos: A lancet. Annals of Plastic Surgery 1989;22:505-515. 2 Mortimer PS. Investigation and management of lymphoedema. Vascular Medicine Review 1990;1:1-20. 3 Kinmoth JB. The lymphatics - diseases, lymphography and surgery, 1st ed. London: Edward Arnold, 1972:87-152. 4 Gamba JL, Silvemaan PM, Ling D et al. Primary lower limb extremity: CT diagnosis. Radiology 1983;149:218. 5 Svensson WE, Mortimer PS, Tohno E et al. The use of colour doppler to define venous abnormalities in the swollen arm following therapy for breast carcinoma. Clinical Radiology 1991;44:249-252. 6 Kissin MW, Querci Della Rovere G, Easton D et al. Risk of lymphoedema following the treatment of breast cancer. British Journal of Surgery 1986;73:580-584. 7 Mortimer PS, Regnard CFB. Lymphostatic disorders. British Medical Journal 1986;293:347-348. 8 Drinker CK, Field ME. The protein content of mammalian lyniph and the relation of lymph to tissue fluid. American Journal o f Physiology 1931;97:32-39. 9 Hadjis NS, Carr DH, Banks L e t al. The role of CT in the diagnosis of primary lymphoedema of the lower limb. American Journal of Roentgenology 1985;144:361 364. 10 Vaughan BF. CT of swollen legs. Clinical Radiology 1990;41:24-30. 11 Stewart G, Hurst PAE, Lea Thomas M e t al. CAT scanning in the management of the lymphoedematous limb. Immunology and Haematology Research 1983;2:241-243. 12 Kiihnke E. Statischer wirksamkeitsnachweisder der manuellen lymphdrainage. In: Leduc A & Lievens P, eds. Lymphokinetics, proceedings o f the international colloqium, Brussels, 1978. Basel: Birkhauser Verolag. 13 MINITAB| Version 7 (1989), Minitab Inc., 3081, Enterprise Drive, State College, PA 16801, USA. 14 Piller NB, Clodius L. The use of tissue tonometry as a diagnostic aid in extremity lymphoedema: A determination of its conservative treatment with benzopyrones. Lymphology 1976;9:127-132. 15 Recommendations of the International Commission on Radiological Protection. ICRP Publication 60. Annals of the ICRP 1990;21:1-201. 16 Veal S. Desquamative interstitial pneumonia. CT findings before and after treatment with steroids. Chest 1988;93:215-217. 17 Duwell S, Hagspiel KD, Zuber J et al. Swollen lower extremity: Role of MR imaging. Radiology 1992; 184:227-231.

9 1995 Btackwell Science Ltd, Clinical Radiology, 50, 541-544.