There are increasing reports of antiretroviral therapy (ART) drug-related kidney dysfunction. Traditional markers of kidney dysfunction such as urine protein/creatinine ratio and estimated glomerular filtration rate (eGFR) have thus far proven ineffective at detecting some sub-clinical forms of ART-related kidney injury. This is a cross-sectional examination of 114 people living with HIV (PLWH), either naïve ( =104) or treatment experienced ( =10). Urinary kidney injury molecule-1 (KIM-1 ng/mg) thresholds were estimated using electrochemiluminescent assays from stored urine samples and normalised for urinary creatinine excretion (KIM-1/Cr). Correlation coefficients and predictors of kidney tubular injury were compared and derived for both adjusted and unadjusted urinary KIM-1/CR (ng/mg). In PLWH (both ART-naïve and treatment experienced) had a higher baseline unadjusted and adjusted median (≥3.7 ng/mg) and upper tertile (≥6.25 ng/mg) urinary KIM-1/Cr levels compared to either non-normal volunteers (0.39 ng/mg) or those with acute kidney injury in the general population (0.57 ng/mg). When upper tertile KIM-1/Cr (≥6.25 ng/mg) was utilised as a marker of kidney injury, eGFR (ml/min/1.73 m), white Caucasian ethnicity, and protease inhibitor exposure were significantly associated with increased risk of kidney injury in multivariate analyses (odds ratio 0.91, confidence interval [CI] 0.68-0.98, P = 0.02; odds ratio 8.9, CI 1.6-48.6, p = 0.01; and odds ratio 0.05, CI 0.03-0.9, p =0.04, respectively). We found a significant degree of sub-clinical kidney injury (high unadjusted and adjusted KIM-1/Cr) in PLWH with normal kidney function (eGFR ≥60 ml/min/1.73 m). We also found a higher baseline KIM-1/Cr (ng/mg) in our study cohort than reported both in normal volunteers and patients with kidney injury in the general population.