Effects of measurement protocols and repetitions on handgrip strength weakness and asymmetry in patients with cancer.

backgroundThe use of handgrip strength (HGS) in clinical cancer research is surging. The association between HGS and outcomes in patients with cancer varied across studies, which might be due to the different measurement protocols for HGS. We aimed to answer three questions: (1) Did the use of various protocols for HGS, along with different numbers of repetitions, lead to significant differences in maximum HGS values? (2) If yes, were these differences clinically significant? (3) Did the differences in HGS protocols and repetitions affect the identification of HGS weakness or HGS asymmetry?

methodsWe continuously recruited adult patients with solid tumours. Two protocols were used to measure HGS: Method A, following the American Society of Hand Therapists guidelines, and Method B, following the National Health and Nutrition Examination Survey guidelines. To analyse HGS, we used the maximal value obtained from either two or three repetitions of the dominant hand or four or six repetitions of both hands.

resultsWe included 497 patients (326 men and 171 women, median age: 58 years). The maximal HGS values, measured with Method B, were significantly higher than those measured by Method A in both men and women, despite repetitions (all P < 0.05). The maximum HGS values were significantly different across the repetition groups, regardless of measurement protocols and sex (all P < 0.01). The protocol-induced differences in maximal HGS values might be clinically meaningful in over 60% of men and 40% of women despite repetitions. The repetition-induced difference was only clinically significant in 4.3-17.8% of men and 4.1-14.6% of women. To identify HGS weakness, using Method A (six repetitions) as the 'gold' standard, the other protocols demonstrated an overall accuracy of 0.923-0.997 in men and 0.965-1 in women. To identify HGS asymmetry, using Method A (six repetitions) as the 'gold' standard, Method B (six repetitions) demonstrated a diagnostic accuracy of 0.972 in men and 0.971 in women. Method A (four repetitions) showed a diagnostic accuracy of 0.837 in men and 0.825 in women, while Method B (four repetitions) showed a diagnostic accuracy of 0.825 in men and 0.807 in women.

conclusionsBoth measurement protocols and repetitions significantly affect the maximal HGS values. The identification of HGS weakness is not significantly affected by either protocols or repetitions, while the identification of HGS asymmetry may be affected by different repetitions but not protocols.