Potential use of the topical niacin skin test in early psychosis—a combined approach using optical reflection spectroscopy and a descriptive rating scale

Abstract

The niacin skin phenomenon reflects a prostaglandin (PG) mediated flush and oedema reaction. As PG metabolism is linked to breakdown of membrane lipids, diminished sensitivity to niacin application suggests potential disturbance in membrane phospholipid—arachidonic acid—PG pathways. We aimed to evaluate and quantify topical niacin skin reaction in early psychosis using optical reflection spectroscopy (ORS) and a new descriptive assessment scale integrating time course, redness, and oedema. Niacin skin tests were performed on 25 medicated first-episode psychosis patients fulfilling DSM-IV criteria for schizophreniform psychosis or schizophrenia and on 25 healthy controls. Nicotinic acid was applied in four dilution steps to the subjects inner forearm skin and skin reaction was consecutively assessed using ORS and a seven point rating scale. Both descriptive ratings and spectroscopic measures revealed significant group differences at the lower niacin concentrations (0.001 and 0.0001 M). At higher concentrations (0.01 and 0.1 M) only descriptive ratings were capable to show significant group effects. Data of both methods showed moderate to strong correlation (r=0.605) as long as the erythema was not affected by the oedema. The data suggest that niacin sensitivity is inversely correlated with negative symptoms. Both methods demonstrate that niacin sensitivity is impaired in a group of first episode psychosis patients and are therefore able to distinguish a subgroup of patients with metabolic impairment. Niacin sensitivity in high risk populations and the specificity of impaired skin response are subjects of further investigation.

Introduction

More than 20 years ago Horrobin introduced a prostaglandin-based pathophysiological concept of schizophrenia, which included the observation, that a varying percentage of schizophrenic patients failed to flush following oral niacin (vitamin B₃) application (Horrobin, 1977, Horrobin, 1980). Substantiated by ³¹Phosphorus magnetic resonance spectroscopic data (Keshavan et al., 1991a, Pettegrew et al., 1991), findings on altered phospholipase A₂ activity (Gattaz et al., 1987, Ross et al., 1997, Ross et al., 1999) and depletion in long-chain fatty acids in erythrocyte membranes and post mortem nervous tissue of schizophrenia patients (Berger et al., 2002b, Fenton et al., 2000) a concept of psychosis based on phospholipid pathology was developed, integrating traditional transmitter-based hypotheses and alterations of membrane metabolism (Horrobin, 1998).

From the first niacin skin tests in 1959 (Murrell and Taylor, 1959) it took more than 30 years to elucidate the link between niacin skin sensitivity and lipid metabolism in more detail. Prostaglandins as striking mediators of this skin phenomenon (Morrow et al., 1992) trigger an increase of intracellular cyclic adenosine monophosphate (cAMP) production, which in turn causes a vasodilatation of superficial skin microvessels (Andersson et al., 1977)—clinically observable as erythema and oedema. The precursors of prostaglandins membrane phospholipids (PGs) (i.e. arachidonic acid) are mainly bound to and released during phospholipid breakdown processes. A depletion of PG precursors leads to disturbed PG production. Therefore, niacin sensitivity gives indirect information about the availability of membrane bound long-chain fatty acids and its bioactive derivates (e.g. prostaglandins).

The notion that schizophrenia comprises a group of disorders with similar clinical picture but different endophenotype (Garver et al., 2000) emphasises the need for markers to select more homogeneous subgroups. The availability of biological markers able to define such endophenotypes, would have major implications not only for diagnosis and treatment, but also as indicators for patients at risk to develop psychosis. The development of an easily applicable topical variant of the niacin skin test (Ward and Glen, 2001) triggered its renaissance in schizophrenia research. Furthermore the recent focus on the unspecific, pre-diagnostic period of schizophrenia and related disorders as one of the most important therapeutic targets (Häfner and an der Heiden, 1999) further supported the search of markers that predate the onset of full-blown psychosis (McGorry et al., 2000). There is increasing evidence, that early intervention in the prodromal phase of disorders is able to prevent or delay the development of a full clinical picture of psychosis, at least in a subgroup of disorders. But the understanding of the underlying mechanism of the transition to psychosis is still limited (McGlashan and Johannessen, 1996, McGorry et al., 1996, McGorry and McConville, 1999). ³¹P-MR-spectroscopic results suggest altered membrane metabolites in first onset schizophrenia patients (Fukuzako et al., 1999, Pettegrew et al., 1991, Stanley et al., 1995), that may be even be present before the onset of psychosis (Keshavan et al., 1997, Keshavan et al., 1991b). Biological markers like the topical niacin skin test may be able to identify subjects with a similar underlying pathophysiology relevant for the implementation of more specific preventive strategies (e.g. supplementation of eicosapentaenoic acid, EPA). Clinical trials on (ω-3)-long-chain fatty acid (i.e. EPA) substitution therapy of schizophrenia patients (Emsley et al., 2002, Fenton et al., 2000, Peet et al., 2001, Peet and Horrobin, 2002) suggest one potential agent targeting a biochemically homogeneous subgroup. Thus, the use of niacin sensitivity as an early marker for increased vulnerability of psychosis and the availability of tolerable treatment options for the early phase of disorder would lead us to new preventive strategies in psychosis treatment.

Cross-sectional studies using the oral niacin flush test revealed heterogeneous results. Those studies that used doses higher than 200 mg niacin and assessed the skin flush response of the upper body visually reported that 25–50% of patients with schizophrenia failed to flush in comparison with normal controls (Glen et al., 1996, Hudson et al., 1997, Rybakowski and Weterle, 1991). Efforts to quantify the assessment of the skin reaction using a photoplethysmograph attached to the ear lobe (Wilson and Douglass, 1986) or malar temperature (Fiedler et al., 1986) demonstrated inconclusive results. Rybakowski and Weterle (1991) performed a methodically combined study using an oral challenge dose of 200 mg niacin and two different techniques (thermometric recordings and visual ratings) for the assessment of niacin response. About 25% of patients with schizophrenia had no clinical flush reaction at all. Though all patients demonstrated a rise in temperature, the time of the maximum temperature following niacin intake was significantly delayed in non-flushing schizophrenic patients. Additionally, the maximal rise in temperature was significantly lower in non-flushing compared with flushing patients with schizophrenia. Although the absolute intensity of skin redness was not recorded, the authors concluded, that flushing may be not an “all or none” phenomenon, but may have a varying degree of intensity. Furthermore the time dimension of niacin response was taken into account for the first time, illustrated by the group differences in time of maximum temperature.

The evaluation of skin temperature changes after oral administration is time consuming and can only be properly performed on fasting patients. Side effects of oral niacin application may be quite unsettling. Such an approach is difficult to implement in clinical routine, especially in the early phase of illness (e.g. drug-naive patients). These considerable disadvantages of the oral niacin test protocol encouraged the development of a topical application of the test (Ward et al., 1998). The test protocol introduced by Ward and colleagues includes the application of four different concentrations (0.1, 0.01, 0.001 and 0.0001 M) on to the inner forearm skin for 1 min. The topical reaction is assessed after 5, 10, 15 and 20 min after removal of the niacin patches. In her pilot study Ward and co-workers (1998) used an ordinate four-point rating scale (0=no response to 3=maximal cutaneous erythema) for the assessment of skin redness and a qualitative description (yes/no) of the oedema. This scale allowed the assessment of the erythema at each time point, but with a limited possibility to integrate the time-course of the reaction and the oedema. In this study all used concentrations revealed a highly significant diminished niacin sensitivity in medicated patients, as compared to healthy control subjects. In a consecutive study Shah and colleagues (2000) integrated the presence of oedema as maximal reaction in the ordinate 0–3 scale developed by Ward et al and could demonstrate that up to 50% of patients with schizophrenia demonstrated an impaired skin reaction at all time points. Berger et al. (Berger et al., 2002a, Berger and McGorry, 2001 developed a descriptive seven-point rating scale integrating erythema, oedema and time-course. About 40% of his first-episode psychosis patient sample had a strong impairment of skin reaction (Berger et al., 2002a).

The current study combined our new method to quantify the erythema using optical reflection spectroscopy (ORS) (Smesny et al., 2001) with the descriptive seven-point scale (Berger and McGorry, 2001). This is also the first study that selected a group of young patients in the early phase of the disorder. Up to now niacin skin test studies were performed in treated (Ward et al., 1998) and untreated (Shah et al., 2000) mainly older multi-episode patients. It is unclear whether metabolic processes underlying niacin sensitivity are especially relevant at the onset of psychosis. The aim of this study was to compare spectroscopic data and descriptive ratings of the same group of acute early psychosis patients fulfilling DSM-IV criteria for schizophreniform psychosis or schizophrenia. This combined approach enabled both the evaluation of the descriptive assessment scale of skin response using ORS and the characterisation of niacin skin response in an early phase of disorder.