Microbiological challenges in the diagnosis of chronic Q fever

Abstract

Diagnosis of chronic Q fever is difficult. PCR and culture lack sensitivity; hence, diagnosis relies mainly on serologic tests using an immunofluorescence assay (IFA). Optimal phase I IgG cutoff titers are debated but are estimated to be between 1:800 and 1:1,600. In patients with proven, probable, or possible chronic Q fever, we studied phase I IgG antibody titers at the time of positive blood PCR, at diagnosis, and at peak levels during chronic Q fever. We evaluated 200 patients, of whom 93 (46.5%) had proven, 51 (25.5%) had probable, and 56 (28.0%) had possible chronic Q fever. Sixty-five percent of proven cases had positive Coxiella burnetii PCR results for blood, which was associated with high phase I IgG. Median phase I IgG titers at diagnosis and peak titers in patients with proven chronic Q fever were significantly higher than those for patients with probable and possible chronic Q fever. The positive predictive values for proven chronic Q fever, compared to possible chronic Q fever, at titers 1:1,024, 1:2,048, 1:4,096, and ≥1:8,192 were 62.2%, 66.7%, 76.5%, and ≥86.2%, respectively. However, sensitivity dropped to <60% when cutoff titers of ≥1:8,192 were used. Although our study demonstrated a strong association between high phase I IgG titers and proven chronic Q fever, increasing the current diagnostic phase I IgG cutoff to >1:1,024 is not recommended due to increased false-negative findings (sensitivity < 60%) and the high morbidity and mortality of untreated chronic Q fever. Our study emphasizes that serologic results are not diagnostic on their own but should always be interpreted in combination with clinical parameters.

Fig 1

Phase I IgG titers at the time of C. burnetii PCR-positive (n = 62) and -negative (n = 25) findings for blood (plasma or serum) of proven chronic Q fever cases. PCR-positive patients include patients with PCR-positive findings in blood with or without positive tissue PCR; PCR-negative patients include patients with negative PCR findings in blood with or without positive tissue PCR results. Logistic regression analysis demonstrated a significant rise in the probability of positive PCR in case of increasing phase I IgG titers using a binary dilution scale (odds ratio [OR], 1.35; 95% confidence interval, 1.02 to 1.77; P = 0.033).

Fig 2

Box plot demonstrating the distribution of titers of IgG antibodies against Coxiella burnetii phase I antigens (phase I IgG) at the time of diagnosis and at the time of peak titers in proven, probable, and possible Q fever cases. The black bars indicate the median IgG titers. Titers on the y axis are represented on a binary scale. Median phase I IgG titers were significantly higher for proven chronic Q fever cases (P < 0.05).

Fig 3

Receiver operator characteristics (ROC) curve of titers of IgG antibodies against Coxiella burnetii phase I antigens at the time of diagnosis and peak titers to differentiate between proven and possible chronic Q fever. The AUC was 0.78 (95% CI, 0.70 to 0.85) for phase I IgG titers the at time of diagnosis and 0.79 (95% CI, 0.72 to 0.86) for phase I IgG peak titers.