Infection control in healthcare settings: perspectives for mfDNA analysis in monitoring sanitation procedures

Healthcare-associated infections (HAIs) in medicine and dentistry are an issue of great concern for public health, as they represent the most frequent adverse effect during care delivery [1]. The global burden of HAIs remains unknown due to the lack of surveillance systems in several countries and to the absence of harmonized criteria for their diagnosis. However, on the basis of the available data, it can be estimated that each year hundreds of millions of patients are affected by HAIs worldwide, with an annual prevalence ranging from 3.5 to 12 % in high-income countries and at least 2–3 fold higher in low or middle-income countries [2].

Dental healthcare settings are associated with a risk of exposure to microorganisms both for dental workers and patients [3‐5]. Microbiological hazards involve a wide number of microorganisms detected in saliva and gingival fluids as well as on contaminated dental instruments [6‐10]. Considering unrecognized or under-reported cases it is assumed that the real threats of cross-transmission in dentistry are probably higher than that of other clinical settings [11]. Techniques for sanitizing reusable equipment are reported as key measures for controlling HAIs in dentistry [6, 12‐18]. These techniques differ according to the reusable items: critical, semi-critical, non-critical items; in particular, critical and heat-tolerant semi-critical equipment should be sterilized by heat (autoclaving, dry heat, unsaturated chemical vapor), heat-sensitive semi-critical equipment should be processed by means of high-level disinfection, and non-critical items should be cleaned and/or disinfected using an hospital disinfectant registered by an official agency such as the United States Environmental Protection Agency (USEPA) [12]. All these procedures should be preceded by a decontamination treatment, applied to reduce residual biological risk for professionals who will perform subsequent treatments of sanitization [18, 19]. However, a lot of research performed in different countries showed that procedures are not harmonized [20]. Traditionally, the effectiveness of disinfection and sterilization protocols is checked by the use of conventional microbiology methods or by other means (i.e. engineering controls) according to the CDC guidelines on infection control [12, 18].

The development and diffusion of molecular techniques, e.g. Real Time PCR, conveyed several advantages in comparison to traditional culture-based methods, being less labor intensive and less time consuming; in addition, they can be tailored to be highly sensitive and specific, at reasonable costs [21]. Nucleic Acid Technologies, not only overcome the restrictions related to classical microbiological tests, but can also avoid the limitations posed by viable but non-culturable cells [22]. Thus, the potential application of molecular techniques represents a challenging opportunity to implement infection control, also in monitoring reprocessed devices exposed to biological fluids.

Recently, the identification and characterization of a biological fluid by the analysis of microflora DNA (mfDNA) has become a key technical approach in forensics [23]. A multiplex real-time PCR assay was developed based on the detection of the microflora genomic signature to identify different human body fluids as salivary, fecal and vaginal fluid [24].

The data from experimental tests performed on dental mirrors contaminated with artificial salivary solutions are summarized in Table 1. All contaminated or partially sanitized samples tested positive for the presence of saliva traces, (C_T < 35, range 20.2 to 29.5). Interestingly, only the samples which were completely processed tested negative. In a series of three replicate experiments, these data were consistent for both red and white solutions. Saliva traces were also identified in the cases of contaminated mirrors that were autoclaved, but not previously cleaned and disinfected. Conversely, fully disinfected and appropriately cleaned samples tested all negative for the presence of saliva traces, further supporting the role of this critical step (data not shown). All the internal negative controls without the addition of bacteria in the artificial saliva solutions were negative.

In full or partial disinfection experiments, when using a lower concentration of disinfectant for a shorter treatment time, the comparison of the amplification curves showed a reduction of about two or three C_T cycle points in comparison to the untreated samples, corresponding to about one magnitude log in bacterial genomic units [26]. However, as shown in Fig. 2, the action of a complete or incomplete disinfectant treatment by itself never provided a negative result after amplification; by contrast, in the culture-based microbiological test, these samples always tested negative. This finding supports the efficacy of disinfection in inhibiting bacterial growth on culture plates (100 %), but also highlights the limitations of traditional microbiology in detecting incomplete sanitation or traces of residual bacteria, as shown by real time PCR.

Sensitivity and linearity limits of the proposed approach are shown in Fig. 3. No significant differences were observed in Red vs White solutions. The lower limit of detection was of 70 fg of template DNA corresponding to the genomic content of about 25–30 cells, in accordance with previous reports [27]. The set of data was characterized by elevated linearity and a correlation coefficient close to 1 (R₂ = 0.99), both for Red and White solutions.

The results from “in field” analyses are reported in Table 2. All the samples correctly sterilized after complete reprocessing resulted clearly negative for Mix_S, as well as for all the other Real Time PCR Mix.

Out of the 22 “in field” used dental mirrors, the Mix_S was able to detect either S. salivarius and S. mutans in 77 %. Only three samples tested positive with the Mix_ES, and one of them was positive for Enterococcus spp. and Staphylococcus aureus. All the samples were negative at the Mix V.

More than half (60 %) of the dental mirrors immersed in the disinfectant bowl were still found to be positive for the Mix_S. However, the remaining 40 % of negative samples were already at a very low level of contamination after use (Ct >35) and only one case testing negative after use became positive (C_T = 27.7) after immersion in the disinfection bowl (sample 4d, see Table 2). In only three cases, the samples were positive for Streptococcus mutans, while one sample was positive at the Mix_ES, and with a double positivity both for E. faecalis and S.aureus Also in this set of the experiment, the samples were all negative for the Mix V.

91 % of the 11 swabs contaminated directly with salivary fluid from human volunteers tested positive with Mix_S, and negative with the other Mix.

Consistently, all the sterilized mirrors were also negative for the presence of human DNA.

Based on the observed results, we calculated the sensitivity, specificity, false positive rate, false negative rate, efficiency and selectivity of the tests. As shown in Table 3, the results were respectively 81 %, 100 %, 0 %, 2, 82 % and −0.11 for “in field” samples when using the raw data (referring to the whole monitoring procedure: sampling-extraction-analysis), and 95 %, 100 %, 0 %, 0.5, 95 % and −0.05 after confirmation of false negative results (referring only to the analysis by a real time amplification method); while for dental mirrors experimentally contaminated with artificial salivary solutions, we obtained the same values for specificity, false positive rate, false negative rate, and selectivity, but observed a further increase in sensitivity (100 %) and efficiency (100 %). Taken together, these data support the effectiveness of the proposed approach.

Increasing knowledge on cross-infection risks in dental healthcare [6, 12‐18] has led to improved surveillance procedures in dental hygiene practice in the last years. Starting from the 1980s, CDC and other agencies have published and updated specific guidelines for reusable instruments, focusing on the cleaning/disinfection/sterilization flow and on the need for effectiveness and appropriateness, to obtain a “step by step” trustworthy sanitation protocol [12, 18‐20]. Implementing the monitoring of sanitization procedures for reprocessing medical devices plays a fundamental role in assuring safety and preventing healthcare-associated infections.

We tested the application of a new molecular approach, based on the identification of residual traces of a biological fluid starting from the detection of its microflora components by mfDNA amplification. In comparison to traditional protocols based on bacteria indicators, this strategy requires an equipped molecular biology laboratory, but seems to carry several advantages. Firstly, it is independent of microbial culture requirements, allowing the detection of traces even after incomplete or unsuccessful reprocessing. Secondly, it is not based on pathogen identification, but on the search for microbial markers whose presence in biological fluids would indicate a possible occurrence of undesirable pathogens, including viruses or prions, thus suggesting a failure in the reprocessing chain. Moreover, the method based on bacterial DNA amplification has the advantage of starting from prokaryotic cells, that are in higher number in saliva in comparison to tissue exfoliated human cells and have a higher DNA resistance to environmental agents. Finally, the saliva microbial signature can be easily implemented based on microbiome achievements and technological advances, representing a promising approach to monitoring sanitation procedures. In comparison to traditional culture based methods, the main limits of this strategy include the availability of a molecular biology laboratory equipped with a real time PCR apparatus, with related costs; moreover, it is not possible to discriminate between live and dead cells; finally, not all the different steps can be evaluated, such as autoclaving performance or partial disinfection. However, it is important to consider that sterilization controls and disinfectant evaluation are already very well established following mandatory rules [18‐20].

We searched for residual salivary traces by mfDNA analysis, in order to evaluate the extent of sanitation of reprocessed dental devices, collected after different treatments. For this purpose, we performed two kinds of evaluation: the Experimental test, based on trials carried out on dental mirrors, experimentally contaminated by two different artificial salivary solutions; and the “In field” test, performed on dental mirrors actually in use in everyday dental care settings. The main result concerns the demonstration of the feasibility of mfDNA analysis as a tool for monitoring sanitation efficacy in reprocessing dental instruments.

In the Experimental test, all the contaminated or partially sanitized samples tested positive for the presence of saliva traces, and only the completely processed samples were confirmed as negative. This result suggests that saliva traces, detected by amplification of mfDNA, can really represent a useful marker for monitoring sanitation. Moreover, the contamination was also identified in samples that were autoclaved without the mandatory cleaning and disinfection steps, further confirming its applicability in surveillance.

Full disinfection and appropriate cleaning represent the fundamental steps for saliva removal, following guidelines [18‐20]. The physical or chemical sterilization step is essential to avoid further contaminations related or not to biological fluids, as well as to safely package and store the reprocessed medical tools [19].

The bacterial genome has a high environmental resistance to chemical and physical stressful conditions. This resistance could affect the differences between full and shorter treatments both for Red and White solutions (Table 1). Data reported in Table 1 show some variability in experiments conducted with Red solution, compared to those performed with White solution, however these differences were not statistically significant.

Interestingly, even after disinfection or autoclave treatment, both white and red solutions were detectable, showing that no amplification inhibition of artificial saliva was induced by hemoglobin or proteins or disinfectant residues, neither after heating or drying steps.

However, it should be emphasized that experimental results were obtained under a controlled situation, without interferences, such as the presence of other microorganisms, environmental agents and using an established load of S. salivarius. For this reason, we also applied the “in field” strategy considering different situations in a blind sampling collection. The proposed approach was successful also when used for “in field” assays, confirming the sanitation of the mirrors correctly reprocessed, and conversely showing the presence of salivary fluid on samples used and not treated, or used but partially or inappropriately processed.

In the “in field” test, we reported a slight reduction in sensitivity and efficiency, mainly due to false negatives resulting from used and not sanitized mirrors. In order to confirm this data we verified samples for the presence of amplifiable DNA, including human saliva traces and cultivable bacteria, excluding inhibitory effects or laboratory cross contaminations. Finally, only one sample out of a total of five proved to be a true false negative; since the mirror was used, DNA was present, human DNA was revealed, but saliva mfDNA was never detected. Another sample, indeed, had no DNA, probably due to a failure in sampling or in the extraction phase or because the mirror was not in contact enough and polluted with saliva. The other three samples showed a relevant bias in the microflora structure since they were collected from patients affected by Candidiasis, suggesting that Candida colonization (or the drug treatment) may have interfered with the levels of S. salivarius in saliva. Even if commensal existence of oral Candida species is not a rare condition [28], its concentration was usually much lower (three folds) than those recovered in these false negative samples. The possibility that a pathology or antibiotics may interfere with this mfDNA-based approach was already considered in forensics applications [29]. This limitation might be easily overcome by increasing the number of samples when monitoring a critical surveillance process. Moreover, the extension of the panel of oral bacteria markers, can further overcome this limit, enhancing the already high level of sensitivity.

The effectiveness of the general principle based on tracing saliva by mfDNA analysis was highlighted by the consistent negativity shown when using another PCR Mix addressed to the identification of other biological fluids (e.g. vaginal and colonic). The sporadic (4 out of 83) observed low positivity for S. aureus or Enterococcus spp. in saliva can be due to accidental contamination, rare but possible [30].

In everyday hospital practice, common disinfection bowls can temporarily join different medical devices, just after their use. When the method was applied to dental mirrors immersed in this container, a possibility of cross contamination emerged, as confirmed also by swab samples collected from the bowl liquid or surfaces, showing positivity for bacteria detectable by Mix_S. Saliva residuals were recovered both on dental mirrors disinfected or sterilized without preliminary cleaning and disinfection. The latter remarks are in accordance with several reports and further highlight the need for following all the required sanitation steps, as recommended in the specific guidelines [6, 14, 18‐20]. This is a critical point that can also be overcome by the mfDNA monitoring [5]. Additional questions may arise for new pathogens or prion proteins that are undetectable by classical methods or could be more resistant to sanitation. This is especially relevant when the infected biological fluids are dried on the glass or metal surfaces of reusable tools, and/or would undergo only partial reprocessing [31]. Traditional methods based on microbial culture or single pathogen detection cannot be applied to monitor most of these events. Molecular tracing of biological fluids may overcome some of these limits related to viable but not cultivable species or due to the detection of microorganisms after disinfection treatments.

The general principle of detecting residual saliva by microflora DNA amplification further shows the multifaceted complexity of monitoring the reprocessing process. The proposed approach based on tracing biological fluids by mfDNA seems to be a promising model and a feasible solution for infection control and prevention in dental healthcare or other hospital settings.

CDC, centers for disease control and prevention; CFU, colony forming units; C_T, cycle threshold; FDA, food and drug administration; HAIs, healthcare-associated infections; mfDNA, microflora DNA; Real-time PCR, real-time polymerase chain reaction