We hypothesized that susceptibility to infectious disease may be a marker of immunodeficiency caused by unrecognized cancer. To test this, we investigated whether there was an excess risk of incident cancer among patients discharged after Staphylococcus aureus bacteremia (SAB) compared with the general population.
Cases of SAB were identified from a national database and linked to the national cancer registry. Controls were matched on age and sex. If a case had cancer before SAB it was excluded including its controls. Incidence rate ratios (IRR) with 95% confidence interval (CI) were calculated by multivariate Poisson regression.
Of 19,088 cases and 170,551 controls, there were 290 (1.5%) and 2286 (1.3%) incident cancers within the first year of discharge, respectively, corresponding to an IRR of 1.77 (95% CI: 1.56-2.02). In the first 90 days, 10 cancers were more likely for cases than for controls (IRR (95% CI) in decreasing order): thyroid (34.69 (2.17-554.64)), testis (16.06 (1.44-179.65)), non-Hodgkins lymphoma (12.40 (4.27-36.04)), liver (9.47 (3.23-27.79)), esophagus (8.89 (2.12-37.20)), urinary (6.34 (2.38-16.86)), kidney (5.17 (1.79-14.93)), breast (3.44 (1.56-7.59)), stomach (3.36 (1.03-10.97)) and prostate (2.13 (1.07-4.25)). Five additional types of cancer were more frequent for cases over the first year: cervix (24.13 (4.84-120.23)), multiple myeloma (7.71 (2.96-20.10)), ovary (7.46 (2.02-27.48)), leukemia (5.13 (2.50-10.51)) and pancreas (2.39 (1.07-5.33)). The IRR of any cancer 2-5 years after SAB was 0.89 (0.80-0.99).
Incident cancer one year after SAB was significantly increased compared to population controls but beyond 2 years the risk was comparable to population controls. 15 cancers were more frequent in cases than controls. Screening for these specific cancers in selected populations may allow for earlier detection.
Fig. 1: Cumulative Hazard Function cases compared to population controls.
N. Mejer, None
R. Skov, None
A. Petersen, None
A. R. Larsen, None
T. Benfield, None