Iatrogenic Meningitis: The Case for Face Masks

Most cases of α-hemolytic streptococcal meningitis after lumbar puncture (LP) are likely caused by dispersal of the organism from the upper airway of medical personnel. Its incidence can be reduced by strict adherence to sterile technique, including wearing a face mask while performing lumbar punctures. Should the wearing of a mask for lumbar punctures be the national standard of care? α-Hemolytic streptococcal meningitis has been reported as a rare complication of LP performed for various reasons—diag-nostic and therapeutic taps, pneumoencephalography, spinal anesthesia (including obstetrical epidurals), and myelography. At least 75 cases (table 1) of meningitis occurring as the result of LP have been reported in the English literature, but the paucity of these reports may not reflect the actual incidence of this infection, which can be expected to rise with the increasing popularity of obstetrical epidurals. In one of the earliest reports, Baumann and Koch [1] present a case and discuss mechanisms by which an organism might gain entrance into the CSF during an LP. It had been shown by Weed et al. [33] that LP performed on experimental animals into which virulent organisms had been recently injected iv resulted almost invariably in meningitis that was due to the injected organism. One hypothesis for the mechanism relates “the sudden lowering of the CSF pressure with a breaking down of the blood-brain barrier” (Baumann and Koch, pg. 1092, [1]). Another hypothesis is that the spinal needle picks up the organism as it punctures small vessels on its path toward the spinal canal and then directly deposits the organism into the CSF. In the reported cases, none of the patients were febrile before the LP, and presumably the initial CSF was clear. This leaves only 2 other mechanisms by which the organism can gain entrance into the CSF—either a contaminated spinal needle was used or the organism gained entrance via an initially sterile needle, either through introduction of contaminated inoculates (e.g., dye, as in myelography, or fentanyl, as in anesthesia) or via droplets from the mouths of medical personnel. The evidence strongly points to droplets from the mouth or upper airway of nearby medical personnel as the source of the organism. The causal organisms, Viridans streptococci, are commensals of the oral cavity and upper respiratory tract. They are also indigenous to the gastrointestinal tract, female genital tract, and skin. The group comprises nonhemolytic or α-hemolytic streptococci, which are not groupable by the Lancefield classification. Usually their specific taxonomic name is used (e.g., Streptococcus salivarius, S. mitis). These organisms are typically of low virulence; when isolated from culture, they are often dismissed as contaminants, but once in spinal fluid they multiply rapidly, causing a full-blown purulent meningitis in 7–24 h. There have been reports of at least 3 instances of clustering—de Jong and Barrs [14] report 2 cases from the same neurologist; Gelfand and Abolnik [20] report 2 cases from the same neurosurgeon; and Schneeberger, Janssen, and Voss [25] report 4 cases from a single anesthesiologist. This is strong evidence that the operator is the source of the pathogenesis, especially considering the rarity of the condition. Clustering is probably related to certain characteristics of the operator (or his or her assistants) that favor the organism's dispersal. Since the organism is present in higher numbers in the presence of dental caries and pharyngitis, these may be factors. The distance of the operator's mouth from the needle's lumen and the extent to which the physician talks during the procedure may also provide an explanation for clustering. Whatever the explanation, the fact that there are clusters suggests an operator-related source. Further support for the droplet hypothesis is provided from The Netherlands, by Veringa, van Belkum, and Schellekens [21]. They identified the source of a case of S. salivarius meningitis after myelography by comparing isolates from the operating neurologist's throat swab and that of his assistant (both grew S. salivarius) with the patient's spinal fluid isolate. The clinicians used polymerase chain reaction (PCR) amplification of extracted DNA and analysis of fatty acid profiles of the organism to discover that the isolate from the neurologist matched the patient's isolate. The assistant's isolate was consistent with the same subspecies but was of a different strain. As a result of this precise identification of the source of the infection, a recommendation was made for medical personnel to wear face masks when performing LP. This recommendation comes at a time when the use of face masks in the operating room is being questioned. This trend is exemplified by Orr [34], who describes an experiment at Severalls Surgical Unit, Colchester, England, where masks were not worn in the operating room for 6 months, during which time the incidence of surgical wound infections did not rise (in fact, it significantly fell). Because α-hemolytic streptococci are ordinarily not wound pathogens in immunocompetent subjects, any opinion about the wearing of masks in the operating room is not applicable in the instance of LP. Here, the CSF provides an excellent culture medium for α-hemolytic streptococci, whereas a surgical wound does not. A good surgical mask (as opposed to a cloth or paper mask) has been shown [35] to effectively eliminate spray from the mouths of talking volunteer anesthesiologists. Oral flora, which include α-hemolytic streptococci, could not be recovered from an agar plate placed at a distance of 30 cm from the mouths of people who were wearing such masks. When the same subjects talked without a mask, organisms grew on >50% of the plates. Excellent protection from an appropriate mask lasts for ∼15 minutes. A proper large, soft, pleated, pliable mask (as opposed to a cloth or paper mask) remains a good bacterial filter for up to 8 h, but it is prudent to change masks for each procedure. Gelfand and Cook [36] defend the absence of masks during the LP procedure, stating, “there are no studies supporting this argument. Given the extreme rarity of the infection, such a study would involve an impractically large number of patients…the efficacy and cost-effectiveness of routinely wearing a mask must be proven” (pg. 131, [36]). As for “extreme rarity,” Gelfand and Abolnik [20] report 3 cases from their community. These cases occurred in a narrow time frame (2 of the cases were caused by the same neurosurgeon). Efficacy of masks to prevent dispersal of the organism has already been demonstrated [35]. Table 1 lists 3 cases that occurred despite the wearing of a mask by the operator. We are given no information about the type of mask worn, how long the operator wore it before and during the procedure, or whether other unmasked personnel were present. Because incidence statistics are not available for Viridans streptococci—related meningitis, it is difficult to arrive at an accurate cost-benefit analysis. Because the cost of a surgical mask is about $0.70 and because this amount is a small fraction of the cost of an LP, further analysis may be superfluous. In cases of meningitis, patients' suffering and attendant financial costs are significant. If one accepts the droplet hypothesis and the concept that a mask can prevent some (if not all) cases of Viridans streptococci—related meningitis, then its benefit justifies its use, despite the rarity of the condition. Some characteristics of reported cases of lumbar puncture-related meningitis