Long-term complications following an outbreak of giardiasis
Not peer reviewed
MetadataShow full item record
Background: The association between certain acute infections and long-term complications is well known, with gastroenteritis and subsequent irritable bowel syndrome (IBS) as one established example. In 2004 there was a large outbreak of Giardia lamblia in Bergen, Norway, due to contaminated drinking-water. An estimated 5000 inhabitants fell ill with giardiasis. Before this outbreak, the knowledge on long-term complications following giardiasis was scarce.
Aims: The overall aim of the studies constituting this thesis was to investigate long-term consequences of having had a Giardia lamblia infection in 2004.
Methods: All three papers in this thesis are reports from a controlled cohort study. In Bergen, Norway, 1252 persons had a verified Giardia lamblia infection by detection of cysts in their stools during the outbreak. These were defined as the exposed population in the study and were matched 2:1 on sex and age to a control group from the Bergen area. Questionnaires were mailed to the participants three, six and ten years after the outbreak. In paper 1 the main outcome was perceived food intolerance and its association with exposure to giardiasis three years after the outbreak. We also investigated the relation with IBS. Perceived food intolerance was measured by two unvalidated questions. IBS was defined by the Rome III criteria. In paper 2 the main outcomes were IBS and chronic fatigue (CF). We investigated the association between giardiasis and IBS/CF ten years later, and changes in prevalence from three to ten and six to ten years. CF was defined by the Fatigue Questionnaire. In paper 3 the main outcome was quality of life (QoL), as measured by the short-form 12 version 2. We investigated the association between giardiasis and QoL ten years later, and further, the relationship with IBS/CF.
Results: Response rates among exposed were 66%, 61% and 50% after three, six and ten years, respectively. Among controls the corresponding numbers were 35%, 36% and 30%. Perceived food intolerance three years after the outbreak was associated with giardiasis, with an adjusted odds ratio (aOR) of 2.00 (95% confidence interval (CI) 1.65 to 2.42), as compared to the control group. Dairy products was the most frequently reported intolerance, with an aOR for exposure of 1.95 (95% CI 1.51 to 2.51). We found no interaction between exposure to giardiasis and IBS on perceived food intolerance in stratified analyses. We found a prevalence of IBS after ten years that was 43% (248/576) among exposed and 14% (94/685) among controls (aOR 4.74; 95% CI: 3.61 to 6.23). For CF the prevalence was 26% (153/587) and 11% (73/692), respectively (aOR 3.01; 95% CI 2.22 to 4.08). There were no changes in the prevalence of IBS among the exposed from six (40%) to ten (43%) years (aOR for the change 1.03; 95% CI: 0.87 to 1.22). The prevalence of CF decreased from 31% to 26% among exposed from six to ten years (aOR for the change 0.74; 95% CI: 0.61 to 0.90). Exposure to giardiasis was associated with a lower QoL. The mean physical component summary T-score among the exposed (51.4; 95% CI: 50.6-52.1) was 2.8 points (95% CI: −3.8 to −1.9; P < 0.001) lower than among controls (54.2; 95% CI: 53.7-54.8). The mean mental component summary T-score was also 2.8 points (95% CI: −3.8 to −1.9, P < 0.001) lower among the exposed (48.9; 95% CI: 48.2-49.6) than among controls (51.7; 95% CI: 51.1-52.4). Adjusting for IBS and CF in regression analyses resulted in no effect of Giardia exposure on the physical component T-score, with an estimated difference of -0.5 points (95% CI: -1.4 to 0.40; P-value: 0.28). Corresponding numbers for the mental component summary in this model were -0.75 (95% CI: −1.7 to 0.22; P-value: 0.13).
Discussion: We found that giardiasis was associated with perceived food intolerance after three years. This is a novel finding. Stratified analyses with IBS and exposure status as independent variables and perceived food intolerance as the outcome indicated a strong association between IBS and food intolerance. The association between IBS and food intolerance is well established, and our findings were relatively consistent with findings from other studies. The strong association between giardiasis and both IBS and CF ten years after the outbreak is surprising and unprecedented in the literature on long-term complications after gastroenteritis. The prevalence of IBS was unchanged from six to ten years, contrary to findings from studies on bacterial gastroenteritis, where post-infectious IBS has been found to subside with time. The lower QoL among exposed than controls was statistically significant, but the clinical significance is questionable. We found no effect of exposure on QoL after adjusting for IBS and CF, indicating that these complications were the basis for the reduced QoL among the exposed. The main methodological problems with our data were the low response rate among the exposed after ten years, and the consistently low response rates among controls, as well as a lack of baseline information about study participants. Analyses were performed to assess selection bias, and the main results from paper 2 would be significant even in the unlikely event of an extreme selection bias. A strength of all the studies was the high number of participants and the inclusion of a control group.
Conclusions: Exposure to Giardia lamblia was associated with long-term complications up to ten years later.