By Carol A. Kemper, MD, FACP
Time for a sea change in infection control
Palmore TN and Henderson DK. Carbapenem-resistant enterobacteriaceae: A call for cultural change. Ann Intern Med 2014; 160(8): 567-569.
This excellent article nicely summarizes the current threat of multi-drug resistant organisms to health care in the United States, detailing the catastrophic effects, both in terms of lives lost, human effort, and dollars spent, because of a single outbreak of carbapenem-resistant infection at the NIH. Only through the use of rigorously applied infection control practices and the use of newer molecular techniques, which could more rapidly drill down on the "gaps" in surveillance and controls, was the outbreak stemmed. As the authors state, the ability to identify these gaps "while the trail is still warm" was essential in stemming the outbreak. Hospitals elsewhere in the U.S. without access to these newer technologies may not be so fortunate.
At our hospital, in an effort to improve compliance with hand cleansing and personal protective gear, there is an ongoing debate with employee and medical staff about the merits of both practices (especially with some of our surgical staff, who not only walk on water, but walk right past towers of gowns and gloves, signage, alcohol dispensers and sinks, without seeing them). The infection control staff has been puzzling through this problem. Agreed, part of the art of medicine is the human touch — the laying of hands, the feel of a hot brow or a tense belly. And through that touch, we hope to heal. Is the experience or impact of that healing touch significantly diminished when wearing gloves ?
One of the problems is that newer surveillance practices are identifying patients who are simply colonized with resistant organisms — they’re not symptomatic from the MRSA in their nares or the ESBL or CRE in their stool, at least not yet. (And their bed, bedside tray table, and sink area are definitely not symptomatic !) One physician recently argued that he used gloves when touching a patient "symptomatic" with C. difficile, saying he needed to protect himself, but completely missing the point. .
There needs to be a sea change in how nursing and physician staff think about patient flora and common contact, with "complete vigilance and scrupulous consistency in adhering to infection prevention principles." Hospitals all over the U.S. are under "siege" from organisms being brought in through their doors, often by patients simply colonized with the newer multi-drug resistant "flora." As much as direct patient contact may be essential to healing, we touch patients all the time who may unknowingly harbor organisms potentially dangerous to others, not just patients with recognized infections or even identified colonization thru enhanced surveillance. Hence, the need to be vigilant all the time with good hand cleansing practices. Even the hands putting on gloves should be cleansed. We can no longer assume that because a patient appears healthy or lacks specific symptoms, they do not harbor multi-drug resistant gram negatives or even C. difficile. Hospital acquired infections do not occur until someone spreads them in hospital.
Does Herpes zoster increase stroke risk?
Langana Sinead M, et al. Risk of stroke following Herpes Zoster: A self-controlled case-series study. Clin Infect Dis 2014; doi 10.1093/cid/ciu098. First published online April 2, 2014.
Using the UK Clinical Practice Database, which includes longitudinal data for nearly 8% of the population of the UK, these authors examined whether a diagnosis of Herpes zoster increased the risk of stroke. Clinical data have suggested an increased risk for stroke within a year following an episode of zoster. Zoster infection may result in invasion of arteries and vasculopathy and even frank vasculitis; certainly the number of patients who exhibit positive CSF PCRs for VZV during episodes of zoster speaks to the ready invasion of the central nervous system. Zoster may also result in a systemic inflammatory response, which could lead to plaque rupture and thrombotic events. One can also imagine the sheer stress and discomfort associated with zoster might result in a hypertensive event.
Adults > 18 years of age with first ever zoster and first ever stroke occurring within a 12-month period between 1987 and 2012 were identified, revealing a total of 11,997 cases. Of these, 6,584 met inclusion criteria for the study. Patients could have no prior history of zoster or stroke; and individuals with a history of TIA, subarachnoid hemorrhage or cerebral aneurysm were excluded. The median age was 77 years, 57% were women, and the median period of observation was 12.5 years. Most cases of zoster did not specify a site of involvement; but 6% specifically indicated zoster opthalmicus and 0.5% had trigeminal nerve involvement. Systemic antiviral therapy was administered to 55%.
The rates of stroke were determined for the period of "exposure", starting the day after the zoster diagnosis and extending to a period of 12 months, divided into 4 time intervals (1-4 weeks, 5-12 weeks, 13-26 weeks, and 27-52 weeks). Overall, compared with baseline, rates of stroke were significantly increased post zoster-event at 1-4 weeks (incidence ratio [IR], 1.63) and at 5-12 weeks (IR, 1.23), and then gradually subsided by 6 months post-diagnosis. A lack of systemic antiviral therapy nearly doubled the observed risk of stroke at 1-4 weeks post-event (IR 2.14) compared with those receiving antivirals. For those receiving antivirals, the risk of stroke, compared with baseline, was significantly increased at 5-12 weeks (IR 1.28) but not for the other time intervals. And, surprisingly, a diagnosis of zoster opthalmicus was strongly associated with an increased risk of stroke at weeks 5-12 post-zoster (IR 3.38). The greatest risk of stroke was observed for persons with zoster opthalmicus who did not receive antivirals, who had > 5fold risk of stroke at 5-12 weeks (IR 5.47). Similar results were observed when combining individuals with ocular and trigeminal involvement.
Examining a large longitudinal population database, an episode of zoster significantly increases the risk of stroke within 6 months, especially when antiviral therapy is not received, or the site of zoster involvement is either ocular or trigeminal. Interestingly, these data suggest a somewhat delayed onset of risk, approximately 5-12 weeks following a diagnosis, which would suggest some kind of delayed affect of viral infection, rather than a direct result of viral replication or invasion of virus into vessels during acute infection. Decreasing the viral load — either in tissues or in the CSF — with antiviral therapy appears important in reducing risk of stroke, even if a delayed risk. Maximizing the use of shingles vaccines, and quickly initiating antiviral therapy are important clinical goals.