Blood profile holds clues to role of infection in a premonitory state for idiopathic parkinsonism and of gastrointestinal infection in established disease

Abstract

The two-stage neuroinflammatory process, containment and progression, proposed to underlie neurodegeneration may predicate on systemic inflammation arising from the gastrointestinal tract. Helicobacter infection has been described as one switch in the pathogenic-circuitry of idiopathic parkinsonism (IP): eradication modifies disease progression and marked deterioration accompanies eradication-failure. Moreover, serum Helicobacter-antibody-profile predicts presence, severity and progression of IP. Slow gastrointestinal-transit precedes IP-diagnosis and becomes increasingly-apparent after, predisposing to small-intestinal bacterial-overgrowth (SIBO). Although IP is well-described as a systemic illness with a long prodrome, there has been no comprehensive overview of the blood profile. Here, it is examined in relation to Helicobacter status and lactulose-hydrogen-breath-testing for SIBO. A robust finding of reduced lymphocyte count in 126 IP-probands and 79 spouses (without clinically-definite IP), compared with that in 381 controls (p < 0.001 in each case), was not explained by Helicobacter-status or breath-hydrogen. This complements a previous report that spouses were 'down-the-pathway' to 'clinically-definite' disease. In 205 other controls without clinically-definite IP, there were strong associations between sporadic cardinal features and immunoglobulin class concentration, not explained by Helicobacter-status. Premonitory states for idiopathic parkinsonism associated with relative lymphopenia, higher serum immunoglobulin concentrations and evidence of enteric-nervous-system damage may prove viral in origin.Although only 8% of the above 79 spouses were urea-breath-test-positive for Helicobacter, all 8 spouses with clinically-definite IP were (p < 0.0001). Transmission of a 'primer' to a Helicobacter-colonised recipient might result in progression to the diagnostic threshold. Twenty-five percent of the 126 probands were seropositive for anti-nuclear autoantibody. In 20 probands, monitored before and serially after anti-Helicobacter therapy, seropositivity marked a severe hypokinetic response (p = 0.03). It may alert to continuing infection, even at low-density. Hyperhomocysteinemia is a risk factor for dementia and depression. Serum homocysteine exceeded the target in 43% of the 126 IP-probands. It was partially explained by serum B12 (12% variance, p < 0.001), but not by Helicobacter-status (gastric-atrophy uncommon in IP) or levodopa treatment. Immune-inflammatory activation increases homocysteine production. Since an estimated 60% of probands are hydrogen-breath-test positive, SIBO, with its increased bacterial utilisation of B12, is a likely cause. Thus, two prognostic indicators in established IP fit with involvement of Helicobacter and SIBO.

Figure 1

Bland-Altman contrast of difference in and average of lymphocyte counts (×10⁹/l) by two methods in 163 subjects. Difference expressed as subset sum (CD3+, CD19+ and CD16+56+)* minus FBC value**. Regression line and 95% upper & lower limits of agreement shown. Difference between subset sum and routine lymphocyte count is small around grand mean for average count, its variability constant over the range. *Four-colour fluorescent cell labelling using the MultiTEST kit in TruCOUNT tubes and a FACSCalibur flow cytometer (Becton Dickinson, San Jose, California, USA). ** Using XE2100, Sysmex UK Ltd., Wymbush.

Figure 2

Distribution of total white cell count, and its decomposition into two Gaussian distributions. Histogram (a) shows log_e transformed raw data. Histograms (b) and (c) are generated from parameter estimates obtained by applying a Gaussian mixture model to log_e (white cell count). They represent the 'best' mixture to replicate the overall distribution: i.e. two distributions tended to provided a better fit (likelihood ratio test, 3 degrees of freedom, χ² = 7.25, p = 0.06), (b) representing 80% of data, (c) a shift-to-the-right. There was no evidence that a mixture of Gaussian distributions provided a better fit to distribution of log_e transformed lymphocyte (age-corrected) or neutrophil counts (χ² = 0.74 & 3.49, p = 0.9 & 0.3, respectively).

Figure 3

Contrast of lymphocyte counts (×10⁹/l) in probands with idiopathic parkinsonism and their spouses with an age-specific 95% reference range based on routine requests. Counts are corrected as if all subjects were male, using the relationship to gender in controls, there being no evidence of an age.gender interaction.

Figure 4

Shifts in percentage distribution of lymphocyte subsets in probands (P), with idiopathic parkinsonism, and their spouses (Ps) compared with reference range (a) mean and (b) its lower and upper reference limits. In P (black bar) and Ps (grey bar), significant differences are denoted by black star and grey star, no subject outside reference limit by black square and grey square. In (b), more than the 2.5% expected had CD4+ (in P), CD8+ (P & Ps) and CD19+ (P) below lower limit (exact binomial test: p ≤ 0.01 in each case), and CD4+ (P & Ps), CD8+ (P) and CD16+56+ (P) above upper limit (p ≤ 0.02).

Figure 5

Lymphocyte and duodenal enterocyte ultrastructure in idiopathic parkinsonism (IP). Electron micrographs shows mitochondria in a representative blood lymphocyte from (a) a man with clinically-definite lP and (b) an age-matched healthy man, and in a representative duodenal enterocyte from (c) this IP-proband (panel to right shows higher magnification) and (d) his spouse. The proband's normal lymphocyte mitochondria are in marked contrast to his and his spouse's long thin duodenal mitochondria. The proband's IP had been diagnosed 7 years previously. He was receiving anti-parkinsonian medication. His spouse had probable-IP on screening. Both had bloating and cyclical diarrhoea going back 10 years. Typically, in the proband, 4 days of unformed stool alternated with constipation in a four-week cycle. In the spouse, explosive watery diarrhoea followed abdominal cramps, over 2 days in 2 week cycles, with normal bowel habit in between. Both had a positive hydrogen-breath-test (criterion: two consecutive values [37] >cut-point of meter manufacturer), and were negative for Helicobacter by UBT, serology, and culture/molecular microbiology on gastric biopsy.

Figure 6

Relationship of serum homocysteine to B12 concentration in idiopathic parkinsonism. Two probands with exceptionally high homocysteine (110 and 135 μmol/l) are excluded: the first had a low B12 but normal folate, the second (with frank H. pylori infection and an empyema) a low folate with a normal B12. Neither had received levodopa.

Figure 7

Estimated odds ratio, with 95% CI, in subjects without clinically-definite parkinsonism, for presence of a given sign, if serum immunoglobulin concentration doubled. Concentrations adjusted to age 60 years and as if all subjects male. Statistical significance of associations with bradykinesia and postural abnormality are, for IgM p = 0.001 & 0.8; for IgG 0.02 & 0.003; for IgA 0.001 & 0.001; for IgA1 0.004 & 0.001; for IgA2 0.2 & 0.07, respectively. Regarding rigidity, presence of activation phenomenon associated with IgA (OR = 1.7 (1.1, 2.6), p = 0.01) and IgA1 (1.6 (1.1, 2.3), p = 0.02).