This prospective study evaluated serum procalcitonin (PCT) and C-reactive protein (CRP) as markers for systemic inflammatory response syndrome (SIRS)/sepsis and mortality in patients with traumatic brain injury and subarachnoid haemorrhage. Sixty-two patients were followed for 7 days. Serum PCT and CRP were measured on days 0, 1, 4, 5, 6 and 7. Seventy-seven per cent of patients with traumatic brain injury and 83% with subarachnoid haemorrhage developed SIRS or sepsis (P=0.75). Baseline PCT and CRP were elevated in 35% and 55% of patients respectively (P=0.03). There was a statistically non-significant step-wise increase in serum PCT levels from no SIRS (0.4±0.6 ng/ml) to SIRS (3.05±9.3 ng/ml) to sepsis (5.5±12.5 ng/ml). A similar trend was noted in baseline PCT in patients with mild (0.06±0.9 ng/ml), moderate (0.8±0.7 ng/ml) and severe head injury (1.2±1.9 ng/ml). Such a gradation was not observed with serum CRP. There was a non-significant trend towards baseline PCT being a better marker of hospital mortality compared with baseline CRP (ROC-AUC 0.56 vs 0.31 respectively). This is the first prospective study to document the high incidence of SIRS in neurosurgical patients. In our study, serum PCT appeared to correlate with severity of traumatic brain injury and mortality. However, it could not reliably distinguish between SIRS and sepsis in this cohort. This is in part because baseline PCT elevation seemed to correlate with severity of injury. Only a small proportion of patients developed sepsis, thus necessitating a larger sample size to demonstrate the diagnostic usefulness of serum PCT as a marker of sepsis. Further clinical trials with larger sample sizes are required to confirm any potential role of PCT as a sepsis and outcome indicator in patients with head injuries or subarachnoid haemorrhage.