The best magazine
Discussion
Nasal swabbing using agar culture identified two-thirds of the total MRSA carriers who were diagnosed by multiple body site screening using agar plus nutrient broth culture. Nasal swab screening combined with culture on agar is a commonly applied method for detecting MRSA carriage. It is, however, costly in terms of staff time and laboratory processing, and the sensitivity of the technique in detecting true carriers in the general patient population is poorly understood. This study sought to determine the likely true sensitivity of nasal swabbing and the effect on ascertainment of swabbing additional body sites.
One of the key findings in this study was that nasal swabbing alone appears to detect only 66% of "true-positive" cases as assessed by the gold-standard measure (all body site swabs on chromogenic agar plus broth culture combined). There is no way of assessing how many additional cases the gold standard may have missed, so the value of 66% for nasal/chromogenic agar screening is a best-case estimate. This efficiency of identifying MRSA carriers will be further reduced in the real-time hospital environment by the documented difficulty in ensuring compliance with swabbing—observed compliance rates during the Pathfinder study ranged from 80% at the outset to 90% during the latter stages (and then only with considerable additional input to maximize compliance in the context of the study). Therefore, a realistic estimate of 80% compliance with universal nasal swabbing would detect only approximately 53% of true MRSA carriage. This suggests that, with a strategy of universal nasal screening, almost half of true MRSA carriers would go undetected in practice.
MRSA colonization was detected most frequently by nasal screening and least frequently by axilla screening. For a 2-swab approach, the combination of nasal plus perineum swabbing produced a significantly better detection rate (82.2%) than nasal swabbing alone (66.4%); nasal plus throat swabbing also produced a better detection rate than nasal swabbing alone (76.5%), but with overlapping confidence intervals. Perineal colonization is a proxy measure for rectal colonization, which is reported as being more likely to cause environmental contamination and has been associated with high dispersal. Perineal swabbing on this basis would be the site of choice for second swab screening given this propensity for transmission; however, it may be less acceptable to patients than throat screening and more demanding of staff time (patients may require assistance to undress and maneuver). Compliance with a universal 2-swab approach may thus be lower than nasal swabbing alone but could potentially be applied more rigorously to selected higher-risk groups.
A broad range of individual-site detection rates are quoted in the literature and are generally higher than those in this study. However, there is generally no gold-standard measure of total colonization other than combined swab/agar results in these other studies, and positive results for swab/agar testing only within this study are broadly similar for nasal positives (72% [245/273] of all swab/agar isolates here, compared with 70.5% and 73.2% elsewhere); findings were similar for nasal plus throat (83.5% vs 82.2%) and nasal plus perineal (90% vs 89.6% and 92.2%). The isolation rates described by other studies can be higher, but these studies vary in their population samples and detection methods. Some studies were undertaken with inpatients at a higher risk of colonization or combined clinical samples with screening samples.
The actual clinical impact on infection rates of relatively inefficient detection of MRSA carriage is unknown, but the Scottish Pathfinder project found indications of reduced MRSA infection and carriage associated with a 1-year period of universal nasal swabbing at an overall 85% compliance. A recent study also suggests that even relatively low ascertainment of MRSA carriage may be effective. The Pathfinder study found that approximately half of the MRSA infections diagnosed in the hospital occur in patients not recognized as being colonized on admission; those patients will be partly undiagnosed carriers and partly true negatives who acquire their colonization or infection directly or indirectly from patients who are colonized at admission. A recent Scottish study that examined the dynamics of MRSA transmission during a program of universal nasal screening and decolonization found the same overall MRSA colonization prevalence on admission and discharge of 2.9%, but it also found that 1.3% of patients who were MRSA positive on discharge had not been positive on admission.
For those admissions with indwelling devices or wounds (surgical wounds, pressure ulcers, diabetic ulcers, etc), MRSA detection is considerably improved by including clinical site swabbing in the screening strategy. Nasal swabbing alone identified only 40% of positive MRSA admissions in this subgroup, but a combination of nasal and wound/device swabbing identified 100% of confirmed carriers. For this small group, therefore, there is no benefit in recommending additional body site swabbing; however, it does reemphasize the need for stringent application of the current UK recommendation on swabbing all clinically significant sites on admission. Notwithstanding this, the simpler practicalities of reliably applying a 2-swab (nasal plus perineal) regimen to all patients probably outweigh the marginal financial benefits of using nasal swabs only in these patients.
The major potential limitation to the validity and generalizability of the findings presented here is the opportunistic nature of recruitment of patients to the study. Previous experience with the Pathfinder study suggests that screening compliance is lower in short-stay patients, who are also likely to have a lower prevalence of MRSA carriage. While this may increase the apparent carriage rate in the hospital population, this will be counterbalanced at least to an extent by the residual true positives who remain undiagnosed. Any systematic bias should also be mitigated by the fact that 2 different hospital types were involved and 2 different operational strategies for screening were employed; for the latter, it was noted that recruitment was more problematic where dedicated screeners rather than ward staff were employed.
Universal nasal swabbing for MRSA is less effective in practice than previously thought in identifying patients with MRSA carriage, but improving ascertainment from 54% to 72% (assuming 80% compliance) by using a 3-fold combination of nasal, throat, and perineal swabs would come at a significant cost in terms of staff time and resources. Further study of the parameters and economic modeling around the various approaches suggested by this study is required to inform national policy options and will be the subject of an additional publication.
Source: ...