What are the major research findings in people with CFS?

• a unique pattern of infection with a recognised or novel pathogen (Straus 1993)
• altered CNS function due to an abnormal immune response against a common antigen (Lloyd et al. 1993; Hickie and Lloyd 1995d)
• a neuroendocrine disturbance (Demitrack et al. 1994)
• a neuropsychiatric disorder with clinical and neurobiological aspects suggesting a link to depressive disorders (Wessely 1993)
• a psychologically determined response to infection or other stimuli occurring in "vulnerable" individuals (Imboden et al. 1961; Abbey 1993).

What are the gaps in our knowledge and priorities for future research?

However, there is good evidence excluding several candidate mechanisms for the disorder, including retroviral infection, neuromuscular disturbance and structural brain damage. The case for other proposed mechanisms, including an immunological, virological, metabolic or neurohormonal disturbance, remains unresolved and warrants further investigation.

The key confounding variables in studies of CFS include the likely heterogeneity within patient groups being studied and the need for standardisation of laboratory methods used in the investigations. Future studies must seek to identify potentially homogeneous patient groups, including subjects collected prospectively from defined infectious or other putative causes of CFS. Control subjects should include both matched healthy individuals and those with other fatigue-related conditions. The laboratory techniques must be sufficiently reliable, standardised and adequately described for the studies to be replicated.

As effective treatments for people with CFS are likely to follow (not precede) delineation of the patholological processes underlying the disorder, considerable research efforts are required to define the pathophysiology. In the meantime, symptomatic drug therapy is likely to remain one of the cornerstones of management, along with the other approaches outlined in Part 4.

Because of the heterogeneity of people with CFS, symptomatic treatment must be individualised. It is therefore difficult to evaluate proposed new therapies of this kind in a systematic way. However, methodology for "N = 1" randomised trials in individual patients has been described, and is well suited for determining optimal therapy in people with CFS (Guyatt et al. 1986).