Background: Resistance training (RT) is an intervention frequently used to improve muscle strength and morphology in old age. However, evidence-based, dose-response relationships regarding specific RT variables (e.g., training period, frequency, intensity, volume) are unclear in healthy old adults.

Objectives: The aims of this systematic review and meta-analysis were to determine the general effects of RT on measures of muscle strength and morphology and to provide dose-response relationships of RT variables through an analysis of randomized controlled trials (RCTs) that could improve muscle strength and morphology in healthy old adults.

Data sources: A computerized, systematic literature search was performed in the electronic databases PubMed, Web of Science, and The Cochrane Library from January 1984 up to June 2015 to identify all RCTs related to RT in healthy old adults.

Study eligibility criteria: The initial search identified 506 studies, with a final yield of 25 studies. Only RCTs that examined the effects of RT in adults with a mean age of 65 and older were included. The 25 studies quantified at least one measure of muscle strength or morphology and sufficiently described training variables (e.g., training period, frequency, volume, intensity).

Study appraisal and synthesis methods: We quantified the overall effects of RT on measures of muscle strength and morphology by computing weighted between-subject standardized mean differences (SMDbs) between intervention and control groups. We analyzed the data for the main outcomes of one-repetition maximum (1RM), maximum voluntary contraction under isometric conditions (MVC), and muscle morphology (i.e., cross-sectional area or volume or thickness of muscles) and assessed the methodological study quality by Physiotherapy Evidence Database (PEDro) scale. Heterogeneity between studies was assessed using I2 and χ2 statistics. A random effects meta-regression was calculated to explain the influence of key training variables on the effectiveness of RT in terms of muscle strength and morphology. For meta-regression, training variables were divided into the following subcategories: volume, intensity, and rest. In addition to meta-regression, dose-response relationships were calculated independently for single training variables (e.g., training frequency).

Results: RT improved muscle strength substantially (mean SMDbs = 1.57; 25 studies), but had small effects on measures of muscle morphology (mean SMDbs = 0.42; nine studies). Specifically, RT produced large effects in both 1RM of upper (mean SMDbs = 1.61; 11 studies) and lower (mean SMDbs = 1.76; 19 studies) extremities and a medium effect in MVC of lower (mean SMDbs = 0.76; four studies) extremities. Results of the meta-regression revealed that the variables “training period” (p = 0.04) and “intensity” (p < 0.01) as well as “total time under tension” (p < 0.01) had significant effects on muscle strength, with the largest effect sizes for the longest training periods (mean SMDbs = 2.34; 50-53 weeks), intensities of 70-79% of the 1RM (mean SMDbs = 1.89), and total time under tension of 6.0 s (mean SMDbs = 3.61). A tendency towards significance was found for rest in between sets (p = 0.06), with 60 s showing the largest effect on muscle strength (mean SMDbs = 4.68; two studies). We also determined the independent effects of the remaining training variables on muscle strength. The following independently computed training variables are most effective in improving measures of muscle strength: a training frequency of two sessions per week (mean SMDbs = 2.13), a training volume of two to three sets per exercise (mean SMDbs = 2.99), seven to nine repetitions per set (mean SMDbs = 1.98), and a rest of 4.0 s between repetitions (SMDbs = 3.72). With regard to measures of muscle morphology, the small number of identified studies allowed us to calculate meta-regression for the subcategory training volume only. No single training volume variable significantly predicted RT effects on measures of muscle morphology. Additional training variables were independently computed to detect the largest effect for the single training variable. A training period of 50-53 weeks, a training frequency of three sessions per week, a training volume of two to three sets per exercise, seven to nine repetitions per set, a training intensity from 51 to 69% of the 1RM, a total time under tension of 6.0 s, a rest of 120 s between sets, and a rest of 2.5 s between repetitions turned out to be most effective.

Limitations: The current results must be interpreted with caution because of the poor overall methodological study quality (mean PEDro score 4.6 points) and the considerable large heterogeneity (I2) = 80%, χ2 = 163.1, df = 32, p < 0.01) for muscle strength. In terms of muscle morphology, our search identified nine studies only, which is why we consider our findings preliminary. While we were able to determine a dose-response relationship based on specific individual training variables with respect to muscle strength and morphology, it was not possible to ascertain any potential interactions between these variables. We recognize the limitation that the results may not represent one general dose-response relationship.

Conclusions: This systematic literature review and meta-analysis confirmed the effectiveness of RT on specific measures of upper and lower extremity muscle strength and muscle morphology in healthy old adults. In addition, we were able to extract dose-response relationships for key training variables (i.e., volume, intensity, rest), informing clinicians and practitioners to design effective RTs for muscle strength and morphology. Training period, intensity, time under tension, and rest in between sets play an important role in improving muscle strength and morphology and should be implemented in exercise training programs targeting healthy old adults. Still, further research is needed to reveal optimal dose-response relationships following RT in healthy as well as mobility limited and/or frail old adults.