Editor,—Infectious agents such asChlamydia pneumoniae orHelicobacter pylori have been linked to ischaemic heart disease (IHD).1 ,2 Raised plasma fibrinogen has been claimed as a possible link betweenH pylori infection and IHD3 ,4; however, fibrinogen as an acute phase protein may only reflect systemic inflammation from other underlying diseases (usually not considered in previous publications).

Our aim was to evaluate retrospectively a possible relation betweenH pylori infection, plasma fibrinogen, and IHD. We then planned to test the hypothesis that raised fibrinogen induced by H pylori is only important in patients with IHD in the absence of other systemic inflammation parameters.5 Finally, we attempted to lower plasma fibrinogen in this subgroup of patients by eradicatingH pylori.

We examined the notes of all patients referred for coronary artery angiography (at least two weeks after myocardial infarction) from August 1994 to August 1995 with suspected or proved IHD for routinely analysed plasma fibrinogen and H pyloristatus. Systemic inflammation was deemed absent if body temperature, leucocytes, C reactive protein (CRP), and erythrocyte sedimentation rate (ESR) were within normal ranges. Subgroups were defined according to H pylori status, IHD, and plasma fibrinogen in the presence or absence of systemic inflammation.

Of 317 patients, 245 (77%) had IHD with a stenosis ⩾ 70% of at least one vessel, 127 (40%) had IHD and H pylori infection. Forty nine of these 127 (15% of all patients) also had raised fibrinogen (> 3.5 g/l). Only 20 of these 49 patients had normal systemic inflammation parameters (6% of all patients). A causal association between IHD and H pyloriinfection was suspected for these patients. This hypothesis was supported by a higher prevalence of raised fibrinogen inH pylori positive patients with IHD in the absence of systemic inflammation (35.1% v17.5%; p = 0.05, one sided χ² test; relative risk (RR) 1.7; odds ratio (OR) 2.0; 95% confidence interval (CI) 0.9 to −4.6). Comparing only patients with increased fibrinogen without systemic inflammation, the prevalence of IHD was higher in H pylori positive patients (95% v63.6%; p < 0.05, one sided χ² test; RR 1.3; OR 7.3; 95% CI 0.7 to −73). By performing multiple regression analysis, serologically determined positive H pyloristatus was significantly (p < 0.001) associated with raised plasma fibrinogen after adjusting for age, history of peptic ulcer, CRP, ESR, and leucocyte count. Bivariable analysis of these parameters influenced fibrinogen levels (but not IHD). Because of their dyspeptic symptoms, 12 of the above mentioned 20 patients agreed to be tested for activeH pylori infection by ¹³C urea breath test and 11 of 12 were positive and included in anH pylori eradication trial described elsewhere,6 10 of them becoming negative. After obtaining written informed consent we had the opportunity to follow up these 10 patients for 6 months (table 1).

In conclusion, H pylori infection may be regarded as a risk factor for IHD in a very small proportion of patients (6%, borderline significance). Our results may explain why even large epidemiological studies do not show a significant association between H pylori infection and IHD.1 Raised plasma fibrinogen could be a link for the development of IHD in this predefined subgroup. Treatment (H pylori eradication) leads to a decrease of plasma fibrinogen in single cases with a known low risk ofH pylori reinfection. However, in the individual patient there is striking evidence to assess fibrinogen repeatedly: spontaneous fluctuations (as seen in two of 10 patients) can possibly reflect the stability or instability of IHD5or interfering concomitant diseases. Before making a decision whether to treat hyperfibrinogenaemia with H pylorieradication, the absence of all other signs of systemic inflammation is essential.