CLINICAL-TRIALS OF DIABETIC NEUROPATHY - PAST, PRESENT, AND FUTURE

This article reviews current knowledge of the etiology of diabetic neuropathy and the outcomes and Limitations of previous trials and discusses future directions for the investigation of its prevention and treatment. Proposed mechanisms for the development of diabetic neuropathy have been widely studied. It has been shown that there is improvement of nerve function associated with some short-term clinical trials of treatments that address a number of possible etiologic pathways. Improvement of morphometry has also been demonstrated in some shortterm clinical trials. However, with the exception of the Diabetes Control and Complications Trial (DCCT), longterm trials with adequate statistical power to evaluate clinical outcome endpoints have not been conducted. The changes in nerve function are similar in most of the clinical trials. For instance, in four clinical trials directed at separate mechanisms (improved glucose control, high myo-inositol diet, therapy with an aldose reductase inhibitor, and therapy with supplementary gamma-linolenic acid), a similar improvement in peroneal motor velocity of 1-2 m/s is observed. This implies that each of the proposed mechanisms contributes equally to the development of neuropathy or that there is some redundancy to their mechanisms. In addition to an etiologic approach, nonspecific neural stimulants, such as gangliosides and nerve growth factors, have also been investigated for the treatment of diabetic neuropathy. with the exception of the prevention of neuropathy by intensive glycemic control, the modest improvements with all other treatments have not led to sufficient evidence to approve any approach. Subanalyses of previous clinical trials suggest that treatment effects are greatest and most clear in patients with mild-to-moderate early neuropathy (stage I or early stage II). Thus, variation in composition and severity of base-line neuropathic disease in past clinical trials may have ''washed out'' any potential treatment effect. It is encouraging that more recent clinical trials have established more rigid inclusion and exclusion criteria so as to recruit only those patients with early or mild-to-moderate disease and a more homogeneous study population. Improvement with treatment has been measured with several markers including nerve conduction velocity, quantitative sensory testing, autonomic function testing, and morphometric changes. Presumably, the combined finding of improved nerve function and improved nerve morphometry will predict improvement in long-term clinical outcomes such as impaired sensation, painful neuropathy, insensitive feet, neurotrophic ulceration, and/or amputation. However, data to support this possibility are still lacking. It is our opinion that the overall design of neuropathy trials must consider present knowledge about complications in general and neuropathy specifically. We recommend that future trials be conducted over long periods with clinically significant outcomes as in the angiotensin-converting enzyme-inhibitor nephropathy trials and the DCCT with the recognition that reducing the development, rather than reversal, of complications is the best that can be reasonably expected. Patients with mild-to-moderate neuropathy (stage I or early stage II) with presumably more metabolic than structural neuropathy would be preferred subjects, and follow-up of 3-5 years is likely to be needed.