Diagnostische Wertigkeit von Untersuchungen zur Diagnostik des sekundären Armlymphödems bei Patientinnen mit Brustkrebs unter besonderer Berücksichtigung der ultrasonographischen Hautschichtendickemessung : Ergebnisse einer Multicenterstudie

Secondary lymphedema is one of the most frequent complications following breast cancer therapy. Scientific research regarding the diagnosis of lymphedema is poor. Development of lymphedema leads to significant long-term side effects, such as decreased quality in nearly all aspects of life. Therefore early diagnosis and treatment are essential. The primary question was if there is sufficient diagnostic quality in high frequency ultrasound cutis measurement (HSCM) and if applicable if it was useful in staging lymphedema. The impact of further clinical and anamnestic factors on the development of arm lymphedema was analysed as a secondary question. In our prospective, consecutive multicentre study, sonographic examinations were performed on both forearms of 274 patients with unilateral breast cancer from 02/05/2006 until 17/01/2007. Primary endpoint in our study was thickness of subcutis which was analysed using Mann-Whitney-U-Test, ROC-curves and Pearson Correlation Coefficient at a 5% level of significance. Further collected variables were analysed by Mann-Whitney-Test and evaluated descriptively. No significant correlation between subcutis difference and arm lymphedema could be established: for the alternative hypothesis in relation to the difference of subcutis as a distinguishing factor, a p-value of 0.135 was calculated. The ROC-curves, with an estimated area under the curve of 0.556 and a 95%-confidence interval (0.481; 0.631), showed a lack of diagnostic quality regarding sensitivity and specificity. Pearson Correlation Coefficient regarding difference of subcutis and difference of circumference of forearm/upper arm was r=0.059 / r=0.133 respectively. The p-value by Mann-Whitney-Test was p= 0.000 for correlation between lymphedema and measurement of circumference both of upper arm and forearm. For the causal relationship between lymphedema and BMI a p-value of p<0.001 was calculated. Self-reported arm and breast symptoms of all patients were collected by questionnaire and showed that there was a significant correlation between reported symptoms and lymphedema (p-value p<0.000 Arm-Score and p<0.001 Breast-Score). Medical history and clinical examination showed that patients with lymphedema of the arm even in stage I have decreased quality of life compared to patients without edema. Working life and social life are also more often restricted. According to our analysis, difference of subcutis thickness is not a favourable parameter for diagnostic distinction. Our study could therefore not confirm promising results by Balzarini et al, Mellor et al and Van der Veen et al. BMI was however identified as a risk factor for development of secondary arm lymphedema. Measurement of arm circumference and self-reported symptoms by patients respectively showed a significant correlation to lymphedema and we recommend incorporating both techniques as screening for lymphedema into general follow-up for patients with breast cancer.