Health & Medical intensive care

Microbial Colonization on Surfaces of Intravascular Devices

Microbial Colonization on Surfaces of Intravascular Devices

Discussion


Our results indicate that the area of heaviest microbial colonization is the proximal segment of the IVDs. This finding was consistent across all concurrently positioned arterial catheters, CVCs, and PICCs. Microbial colonization was heaviest on the external surfaces of IVDs, suggesting that the devices are colonized down from the proximal end of the catheter toward the distal tip, possibly originating from the skin around the IVD insertion site.

Our finding that microbial growth was heaviest on the surfaces of the proximal intravascular segment of the IVDs brings into question the current reliance on culturing only the tip of a catheter to diagnose CRBSI, a choice based on the apparent assumption that regardless of microbial colonization of different parts of a catheter, the risk for blood-borne microbial dissemination does not begin to increase markedly until colonization at the tip reaches a level of more than 15 CFUs/mL. The number of CRBSIs in our study was insufficient to confirm this assumption. Our results indicated that the distal tip of an IVD may not have microbial colonization even when the proximal segment is colonized. Possibly, tests for diagnosis of CRBSI should include culturing the proximal intravascular end of a catheter rather than, or in addition to, just the distal tip. Our finding of a gradient of high to low levels of microbial colonization from the proximal to the distal ends of the catheters has implications for the assessment of risk for CRBSI and may explain the equivocal association of CRBSI with catheter tips in some previous studies. Further clinical epidemiological research is required to establish the relative usefulness of growth of microorganisms on cultures of catheter tips compared with growth on cultures of proximal segments of a catheter as a risk indicator for CRBSI. So far as we know, no study has been published in which the investigators examined entire IVDs so microbial growth in each different segment of each IVD could be compared with growth in the other segments and with other IVDs concurrently placed in patients (Figure 4).



(Enlarge Image)



Figure 4.



Intravascular segments of the intravenous devices that were cultured.





In our study, all IVDs became more susceptible to microbial colonization the longer they were in place. Contrary to the findings of an earlier study, PICCs had a relatively lower risk of colonization than did CVCs. PICCs are usually inserted in the cubital fossa or brachial regions, which have less skin flora than do the jugular and femoral sites used for CVC insertions. The proximity of the jugular site to oral and nasopharyngeal secretions makes that site more susceptible to exposure to microorganisms in these regions. Similarly, the femoral site is more susceptible than other sites to the skin flora of the perianal and genitourinary regions.

Arterial catheters had relatively lower colonization rates than did CVCs but higher colonization rates than did PICCs, supporting the results of our earlier study and the findings of other investigators that when systemic sepsis is suspected in a critically ill patient, the arterial catheter should be accorded the same degree of importance as the CVC.

Our study reinforces the observation that microbial colonization of CVCs and PICCs increases directly as the number or lumens increases. We found that 5-lumen CVCs had a higher risk for colonization than did 4- or 3-lumen CVCs. Antimicrobial-coated CVCs had lower rates of microbial colonization than did non–antimicrobial-coated 4- or 5-lumen CVCs but still had colonization rates marginally higher than did non–antimicrobial-coated 3-lumen CVCs. This finding suggests that the use of antimicrobial-coated CVCs may be of some benefit in reducing the incidence of colonization, especially in immunocompromised patients or patients with early signs and symptoms of sepsis. When 2-lumen PICCs were compared with single-lumen Groshong PICCs, a similar trend was observed. In our ICU, multilumen CVCs are essential for the administration of antibiotics, inotropes, and sedatives. However, after patients have survived the critical episode of an ICU stay, they tend to be transferred to a general unit with the initial multilumen CVC still in place, thereby incurring the risk of colonization or CRBSI. On the basis of our study and other studies, we recommend that such patients have their CVCs removed and alternative intravenous access sought. Selection of an appropriate IVD should be made with respect to treatment options for the patient and the type and length of intravenous therapy required. For example, if a patient requires long-term intravenous therapy, insertion of a PICC would be more appropriate than insertion of a CVC. If longer term vascular access is required, then surgically implanted vascular devices should be used.

Currently, no gold standard exists for the method of obtaining samples from IVDs for microbial testing. We had to use a microbial culture technique that would account for all microorganisms colonizing both the external and the internal surfaces of the IVDs. We used the roll-plate technique of Maki et al because it is arguably still the most common laboratory technique used to measure colonization of microorganisms on the external surfaces of IVDs. For the internal lumens of the IVDs, we used a sterile wire to extrapolate microbial growth on the internal surfaces of each lumen of an IVD without contamination from microorganisms growing on the external surface. The wire was preferable to an endo-luminal brush because the latter would require a different culture technique (ie, in broth), whereas the wire could be applied on the same culture media used in the roll-plate technique, maintaining the rigor and validity of our study.

The microbiological techniques we used allowed comparisons between the different segments and between the internal and external surfaces without contamination from one surface to another in the same catheter and comparisons between different catheters. The techniques we used may not be optimal in the diagnosis of CRBSI, but diagnosis of CRBSI was not the objective of our study. Earlier laboratory evaluation of the sterile-wire technique with serial dilutions of known microbial concentration revealed that this technique was a reliable method for evaluating microbial density.

Because the patient case mix of the hospital resulted in only a single patient in the pediatric unit with a short-term IVD in place for more than 9 days, our results are applicable only to adult patients. Further, the sample size was planned to be adequate to study the primary hypothesis involving comparison of the different segments of catheter of each IVD with itself, including separate examination of CVCs, arterial catheters, and PICCs. The size was not adequate to provide reliable subgroup comparisons between different groups of patients (eg, differences in age, illness severity, disease groups) or different IVDs (eg, arterial catheters vs CVCs or PICCs). Therefore, we regard our analyses of these secondary hypotheses as hypothesis generating rather than hypothesis testing. Some arterial catheters were short, allowing only for a proximal and a distal segment to be compared for microbial growth. Nevertheless, even with the limited numbers of arterial catheters available, we were still able to establish that microbial colonization differed between the various segments of arterial catheters, with heavier growth on the proximal segments.

You might also like on "Health & Medical"

Leave a reply