This week I was consulted to see a patient that was, for an infectious disease physician, fairly routine: an injection drug user with a fever. However, this injection drug user had some particular habits that were interesting and the key to her diagnosis.
She had the usual fever, chills, and muscle aches that are characteristic of a bloodstream infection. She also complained of back pain leading to the suspicion that the infection could have seeded her vertebral column. She had a markedly elevated white blood cell count suggesting a large immune response had been immobilized. An MRI, blood cultures, and echocardiography all ensued. The patient, however, rapidly improved from her initial state and all these studies turned up no answers.
The patient's peculiar behavior I referenced above (and it might not be all that peculiar) was that she, when out of heroin, would attempt to draw up any remaining heroin from cotton she had used in the past to draw up heroin. Cotton is used as a crude filter and her hope was that small remnants of heroin might remain in the cotton and be dislodged into her body if she injected through it. This reminds me of a David Sedaris short story in which he is scouring his carpet for crystal methamphetamine specks. Despite using clean needles, there are other risks that injection drug users often fail to appreciate.
She had cotton fever. Cotton fever is the result of a bacterial toxin produced by the bacteria Pantoea agglomerans which colonizes cotton plants. What occurs during cotton fever is that residuals of the endotoxin are injected along with the heroin remnants. The endotoxin stimulates a substantial immune response triggering intense symptoms but no bacteria is found as this is pure toxin-mediated. Of course, blood cultures and antibiotics should be administered initially as it is difficult to predict if viable bacteria were also injected along with the endotoxin.
Injection drug users present myriad challenges, cotton fever may be one of the more interesting.