Background and purpose: The measurement of tumour oxygenation using Eppendorf oxygen-sensitive needle electrodes can provide prognostic information but the method is limited to accessible tumours that are suitable for electrode insertion. In this paper the aim was to study the relationship between such physiological measurements of tumour hypoxia and the labelling of tumours with the hypoxia-specific marker pimonidazole. Materials and methods: Assessment of tumour oxygen partial pressure (pO2) using an Eppendorf pO2 histograph and immunohistochemical pimonidazole labelling was carried out in 86 patients with primary cervix carcinomas. Pimonidazole was given as a single injection (0.5 g/m2 i.v.) and 10-24 h later pO2 measurements were made and biopsies taken. Tumour oxygenation status was evaluated as the median tumour pO2 and the fraction of pO2 values ≤10 mmHg (HP10), ≤5 mmHg (HP5) and ≤2.5 mmHg (HP2.5). Hypoxia was detected by immunohistochemistry using monoclonal antibodies directed against reductively activated pimonidazole. Pimonidazole binding was scored using a light microscope. Each tumour was evaluated by the relative area pimonidazole at highest score and the accumulated area of pimonidazole labelling from score 1 to 4. Necrosis was measured in HE stained sections. Results and conclusions: The degree of hypoxia assessed by either pimonidazole binding or invasive electrode measurements varied significantly between tumours. There was a trend that the most hypoxic tumours measured by oxygen electrodes had the highest score of necrosis, and no or little pimonidazole binding. However, this observation was not consistent and there was no correlation between pimonidazole staining expressed in this way and oxygen electrode measurements of hypoxia. © 2003 Elsevier Science Ireland Ltd. All rights reserved.