BACKGROUND Neuron-specific enolase (NSE) and S100B protein are brain-origin proteins commonly described to assess the presence and severity of neurological injury. To date, there are limited data examining the influence of scuba diving on these biomarkers, particularly when symptoms of decompression sickness (DCS) occur. The purpose of this controlled study was to determine whether these serum neurochemical markers could be used as 1) indicators of neurological DCS and 2) predictors of incomplete recovery. METHODS Fifty-nine divers with neurological DCS and 37 asymptomatic divers admitted for inadequate decompression, serving as controls, were consecutively enrolled between 2010 and 2012. Blood samples were collected at initial presentation up to 6 hours after dive completion (controls) or onset of symptoms (DCS divers). Biomarkers were quantified in nonhaemolysed samples only. Clinical outcome was assessed at 6 months post-injury. RESULTS The two groups did not differ regarding the variables examined, except for the total dive time which was slightly shorter in the control group. NSE, but not S100B protein, was higher in the DCS group than in controls (P < 0.0001). An NSE level > 15.9 µg L⁻¹ determined by ROC analysis predicted DCS development with a specificity of 100% (95% confidence interval (CI) 90 to 100) and a sensitivity of 24% (95% CI 14 to 36). There was a trend towards a higher likelihood of residual neurological deficits above this cut-off value (P = 0.08). CONCLUSIONS Early determination of NSE was found to be useful for the diagnosis of neurological DCS with a high specificity. However, its clinical applicability in decision making for determining treatment as well as its prognostic value remains to be established. Reliability of S100B protein was not demonstrated in the present study.