Increased intestinal leakiness and associated systemic inflammation are potential contributors to osteoarthritis (OA) and postural imbalance in the geriatric population. To date, no successful treatment to correct postural imbalance in OA is known. We aimed to explore the effects of a multistrain probiotic upon postural imbalance in OA-affected patients. In this randomised, double-blind trial with a placebo group, 147 patients suffering from knee OA (age span = 64–75 years) were divided into placebo (n 75) and probiotics (n 72) study groups. Vivomix 112 billion, multistrain probiotic was given once a day for 12 weeks. The outcomes of study variables were determined first at baseline and later after 12 weeks of intervention. These were Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC), knee flexion range of motion (ROM), pain intensity by visual analogue scale, handgrip strength (HGS), gait speed and balance control assessed in standing, semi-tandem and tandem stances. We determined plasma zonulin to determine intestinal leak along with c-reactive protein and 8-isoprostanes levels. A total of 136 OA patients taking placebo (n 71) and probiotics (n 65) were analysed. The probiotics group exhibited a reduction in pain intensity, disease severity and WOMAC scores along with improvement in balance scores, HGS and walking speed (P < 0·05 for all), no change in ROM, resting pain and 8-isoprostanes levels. The correlation analysis revealed a robust association of balance scores with plasma markers of intestinal leakiness and inflammation in probiotics but not in the placebo group. Probiotics reduce postural imbalance in OA patients partly due to a reduction in intestinal leakiness.