Impact of daily octenidine skin washing on antiseptic tolerance of coagulase-negative staphylococci in two neonatal intensive care units with different bathing practices

Coagulase Negative Staphylococci (CoNS) are responsibl e for 80-90 % of catheter related sepsis in neonates which can cause life threatening and damaging effec ts. NICUs within the UK use various practic es to decolonise neonate s to prevent infection ranging from regular full body bathing to localised skin decontamination before insertion of indwelling devices. There i s a disparity in bathing prac tices for infants admitted onto neonatal units, with some choosing to regularly ba the infants and others not, and some routinely washing with skin anti septics, and others not. Aim: To compare the abundance of CoNS within two UK NICUs with different approaches to skin bathing and to test their toleranc es to antiseptics.

A collection of CoNS from two UK based NICUs with differing bathing routines for neona te s were collated and tested for susceptibility to the antiseptics in use , octenidine and chlorhexidine. Findings: Regul ar bathing of neonates in octenidine did not dec rease the abunda nce of organisms on neona tal skin. Isolates from the unit where octe nidine was in frequent use did not show any increased antiseptic tolerance. Isolates from the unit where regular bathing was not routine practice were less susceptible to both antiseptics.

Frequent whole-body skin washing with octenidine does not appear to result in a lasting reduction in numbers of organisms found on the skin but also does not appear to select for antiseptic tolerant CoNS. . CC-BY 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted June 25, 2022. ; https://doi.org/10.1101/2022.06.23.22276813doi: medRxiv preprint 4

Infection is common amongst premature and low birth weight infants, due to the immaturity of their i m m u n e s ys t e m , s k i n a n d m u c o s a l b a r r i e r s [ 1 , 2 ] . L a t e o n s e t i n f e c t i o n ( L O I ) , o c c u r r i n g a f t e r t h e f i r s t 72 hours after birth, is usually nosocomial and caused by organisms from the skin microbiota or hospital environment [3]. Within ne onatal intensive care units (NICUs), invasive procedure s are often essential for management but i ndwelling catheters a re a major source of infection [4]. Coagulase-negative staphylococci (CoNS) are common skin commensals, which cause up to 80%– 90% of LOI in NICUs. Catheter-related sepsi s can be life-threatening and c ause permanent lifelong injury and disability in survivors, including cerebral palsy and other adverse neurodevelopmental problems [3, 5-8]. CoNS rapidly colonise the skin of infants after birth, with the most prevale nt species being S. epi derm idi s, S . h ae moly t i cus and S. war n e r i [9]. Disruption of the skin barrier by implantation of intrava scular devices can lead to contamination of the outside of these devices (such as central venous catheters [CVC]). This can then lead to bloodstream and ca the te r-related infections, which can in turn lead to systemic infection and neonatal sepsis [4, 10, 11]. Antiseptics are used pre-implantation to minimise the risk of infection at the site of a skin breach. In addition, within both adult and paediatric populations, there is evidence that re gular bathing using antiseptics including chlorhexidine gluc onate (CHX) can reduce the number of hospital acquired infections within intensive care [12-14]. However this has not been observed for CHX-ba sed body washing in neonates[15]. Whilst there are national evidence-based guidelines for antiseptic use in children, there is no UK guidance in place for infants who are less than two months old [16]. Due to this lack of UK standardised guidance for topical antiseptic use within NICUs, there are a large range of practices in operation, from regular full body bathing to just local site decontamination before insertion of indwelling device s[17]. The re is also a wide disparity between different anti septic s and frequency of bathing, with the most common anti septic s used being octenidi ne (OCT), povidone . CC-BY 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted June 25, 2022. ; https://doi.org/10.1101/2022.06.23.22276813doi: medRxiv preprint 5 iodine and alcohol-/aqueous-based CHX which is frequently u sed in widely varying concentrations depending on hospital protocols [18, 19]. CHX is a ca tionic bisguanide with a broad spectrum of antimicrobial ac tivity [20]. It has been shown that regular bathing with CHX significantly reduc es the bacterial skin burden in neonate s [21], however, the duration of this reduc tion and subsequ ent impact on reducing neonatal bloodstream infections and sepsis, is much le ss clearcut [8, 22, 23]. OCT is a bis-pyridine compound which also has a broad spectrum of antimicrobial activity. Very few studies have examined the use of OCT within a neona tal population, however there is evidence that it is e ffective at reducing hospital acquire d infection among st adults and older children [24-26]. OCT ha s been introduced as a body wash as it is reportedly mild and suitable for patients with vulnerable skin. In this work, we compared antiseptic susceptibility of isolates of CoNS from two UK NICUs; one unit carrie s out regular whole-body washes for infants, using an OCT based anti septic (Bradford Royal Infirmary). The other NICU (Norfolk and Norwich Universi ty Hospital) does not routinely ba the infants betwe en admi ssion and discharge. Our hypothesis wa s that CoNS isolates from the skin in infants who undergo daily whole-body skin was hing with OCT will show higher MIC to OCT compar ed to infants who wer e not r outinely bathed. Thus primary outcome should be the aim to determine whether routine washing of babies with OCT impacted the abundance of CoNS isolated from skin, and impacted tolerance to OCT. Secondary aim w as t o s ee i f re g ul a r w as h i n g o f ba b i es w i t h O C T i m p ac te d t o le r a n ce t o C H X c om pa r e d w i t h i n Co N S isolates from infants who were not routinely bathed. Aims : To determine whe the r routine w ashing of babies with OCT may impact the abundance of CoNS isolated from skin, and tolerance to OCT or CHX. . CC-BY 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted June 25, 2022. ; https://doi.org/10.1101/2022.06.23.22276813doi: medRxiv preprint 6

s and Methods Study Sites a nd r o ut i ne c lean sin g p r ac tic es This work involved NICUs in two hospitals; at the Norfolk and Norwich University Hospital (NNUH) bathing or whole-body skin cleansing was not practise d routinely on infants. At the Bradford Royal Infirmary (BRI), infants ≥ 26 weeks we re washed daily using Octenisan® (0.3 % octenidine) which was applied to the skin using cotton wool and then washed off water. Infants <26 weeks or with broken/ immature skin were excluded from the regime a s per hospital policy. Infants ≥27 weeks correc te d gestationl age were eligible for inclusion in this study. Both ce ntres routinely used locally-applied CHX-ba sed antiseptic s (0.015% – 2% CHX in 70% isopropanolol) for pre-procedural skin disin fection be fore the insertion of indwelling catheters. Is olat e c ollec tion As part of a previou s surveillance study from this laboratory in December 2017 to March 2018[27], a panel of ~800 CoNS were isolated from skin swabs taken at the NICU of the Norfolk and Norwich University Hospital (NNUH). Swabs were taken on admission and once week ly from each baby throughout their NICU stay from various body site s including the ear, axilla, groin and rectum. Infants admitted to the BRI NICU also had skin swabs taken on admission and then once weekly for their duration of stay, over a period of 8 weeks (between January and March 2020). All infants admitted to the BRI NICU were eligible for inclusion, regardless of gestational age or expected duration of stay. A single charcoal swab (Amies Charcoal Transport Swab) was used to take a body sweep, incorporating the ear, neck, an axilla, umbilical are a and groin. The swabbing was typically carried out 12-16 hours after wa shing occurred. Swabs were stored locally at 4 o C. Batches were securely packaged and posted to the Quadram Institute Bioscience (QIB), Norw ich, every 3 weeks, where they were stored a t 4 o C upon arrival. . CC-BY 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted June 25, 2022. ; https://doi.org/10.1101/2022.06.23.22276813doi: medRxiv preprint 7 A unique study ID was allotted to each infant enrolled using their anonymised code generated by the BadgerNet neonatal platform (CleverMed, UK). Birth weights, dates of admission and swabs, birth gestational age, gender of infant, birthing method, location of birth and corrected gestational age at enrolment wa s coll ected. No identifying data were transmitted out of the participating sites and completed anonymised data were collated at QIB into a ma ster database. Is olati on o f Co N S Charcoal swabs were streaked on Columbia Blood Agar (CBA; Oxoid Thermo Fiser Scientific, USA), candidate CoNS were then sub-cultured on Mannitol-Salt Agar (MS A; Oxoid Thermo Fiser Scientific , USA). Isolates were tested for coagul ase (Coagulase Test Slides, Millipore, Sigma), and any isolate s suspected to be En teroc occ i were grown on Bile Aesculin Agar (Oxoid Thermo Fise r Scientific , USA). Finally, catalase te sts were used with 20 % hydrogen peroxide. Isolates considered to be CoNS base d on the phenotyping above we re saved and given a unique study number. Anti mic r o bial su scepti bi lity te st ing The minimum inhibitory concentration (MIC) of OCT and CHX was determined for all isolate s according to European Committee on Antimicrobial Susceptibility Testing (EUCAST) guidelines[28]. Muller Hinton (MH) Agar (Oxoid) was prepared with concentrations o f antiseptic s ranging from 0.25 µg/mL to 64 µg/mL. Overnight cultures grown in MH broth were diluted 1/10,000 and 1 µL drops were plated on to the antiseptic containing MH Agar and incubated at 37 o C for 24 hrs. Two control strains, TW20 and F77, were used throughout[29 ]. An MIC breakpoint of 4 µg/mL has bee n suggested to de termine CHX resistanc e; no breakpoints have been proposed for OCT to date although 2 µg /mL has been used previously a s an epidemiological cut off [30]. Stati stics Data were analysed using GraphPad (PRISM 5). Correlation analysis used nonpa rametric Spea rman tests, one-tailed with confidence levels of 95 %. The nonparametric one-tailed T-Test and the Mann- . CC-BY 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted June 25, 2022. ; https://doi.org/10.1101/2022.06.23.22276813doi: medRxiv preprint 8 Wh i t n e y te st w e r e us e d to i d en t if y s ig n ifi ca n t d i ff e re n c es b et we e n M IC d a ta wi t h a 9 5 % c on f i d e n c e level. Ethic s The study protocol was reviewed by the Research Services Manager of the Norfolk and Norwich University Hospitals NHS Foundation Trust and was approved as a surveillanc e study that did not require a formal ethics committee review. . CC-BY 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted June 25, 2022. ; https://doi.org/10.1101/2022.06.23.22276813doi: medRxiv preprint 9

Is olati on o f Co N S from Br ad ford neo n ate s A total of 56 infants from BRI were enrolled in the study. From these infants, 200 skin swabs were transported to QIB. These were made up of both admission swabs (n=30) and swabs taken weekly (n=170). One swab was discarded due to inadequate labelling. Afte r swabs were incubated plates typically demonstrated heavy growth of both Gram-p ositive and Gram-negative isolates demonstrating various colony morphologies. This contra sts with the results in our previou s study from NNUH where individual swabs generally had a smaller number of isolates with less mixed cultures. Of 180 Gram-positive isolates, a total of 78 were confirmed a s CoNS and retained for phenotypic testing. Susc epti bility of Br ad for d CoNS is ola te s t o anti se ptic s The MICs of CHX and OCT were de te rmined for the isolates from BRI. Isolates were generally very sensi tive to OCT and the MICs ranged betwee n ≤0.125-1 µg/mL with the majority (48.7%) of the isolates (n= 38) being inhibited by ≤0.125 µg/mL. Two isolates had a MIC of 0.25 µg/mL, 32 isolates (41%) had a MIC of 0.5 µg /mL and the re maining 6 (7.7%) had a MIC of OCT of 1 µg/mL. The isolates were also tested against CHX. For 44.9 % of the isol ates (n=35) the MIC of CHX wa s ≤0.125 µg/mL, for 29 (37.2% ) it was 0.25 µg/mL, 1 isola te had a CHX MIC of 0.5 µg/mL and the remaining 12 (15.4 %) had a CHX MIC of 1 µg/mL. No isolates from BRI was above the proposed breakpoints for eith er antiseptic. The CHX and OCT MIC data for each of the Bradford isolates were compared against each other to d e t e r m i n e w h e t h e r t h e r e w a s a n y r e l a t i o n s h i p b e t w e e n t h e s u s c e p t i b i l i t y t o t h e t w o a g e n t s . T h i s analysis (Figure 1) showed no direct relationship between susceptibility to the two antiseptics ( P = 0.4), which is similar to our previous findings[29]. Comp arati v e an ti sep tic s u s c epti bility o f is olate s fr o m Br ad ford an d Norwic h . CC-BY 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted June 25, 2022. ; https://doi.org/10.1101/2022.06.23.22276813doi: medRxiv preprint 10 Antiseptic susceptibility of the BRI isola tes from the daily OCT -washed babies was compared with the panel of 863 CoNS isolated from babies in the NNUH NICU where infants are not routinely bathed with OCT. A comparison in the susceptibility profile s of the population of CoNS from BRI and NNUH showed significantly decreased susceptibility in the NNUH population to both antiseptics (figures 2 and 3). The MICs of OCT for infants from NNUH ranged between 1 and 16 µg/mL (mean of 2.319 SEM± 0.078 µg/mL), compared with a narrower range of ≤ 0.125 and 1 µg/mL (mea n of 0.394 SEM± 0.029 µg/mL) for BRI isol ates (Figure 2a). There was a significant difference in the mean MIC for OCT between Bradford and Norwich NICUs, ( P= <0.0001, Figure 2B). The MICs of CHX for NNUH isola tes ranged between 2 to 64 µg/mL (mean of 20.1 SEM± 0.5 µg/mL), compared with a range of ≤ 0.125 to 1.0 µg/mL for isolates from babies at BRI (mean of 0.31 SEM± 0.04 µg/mL) (Figure 3). A clear difference in the distribution of CHX susceptibili ty of the isolates from the two sites can be observed. There was a significant difference between the mean MIC for CHX between the Br adfor d and N NUH isolates ( P= 1 µg /mL. . CC-BY 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted June 25, 2022. ; https://doi.org/10.1101/2022.06.23.22276813doi: medRxiv preprint 11 Dis c us sion In this study we sought to examine the prevalence and antiseptic susceptibility of CoNS from the skin of ne onates where washing with OCT is routine daily practice, and to compare the antiseptic susceptibility with a panel of isolate s from a unit which does not bathe infants routinely in the period between admission and discharge. Inoculation of swabs from BRI typically resulted in extensive bac terial growth made up of multiple morphologically distinct bacteria, including a mixture of Gram-positive and Gram-negative bacteria . Therefore, despite being bathed regularl y using OCT, the numbers of organisms on the skin of babies from BRI remained high, and in fact more variation was se en than for swabs attained from the NNUH using the same methodology, althoug h this is an anecdotal observation. T he skin swabbing technique in BRI involved carrying out a whole-body composite swab, which incorporated the ear, neck, axilla, umbilical area and groin. It has been suggested that there is not much differentiation between differing sites of the body and the skin burden, however this may have c ontributed towards the high number of organisms recovere d, and in particular the greate r numbers of puta tive Gram- negative bacteria [31]. Swabbing was carried out up to 16 hours after the wa shing of the infants occurred, sampling time was not standardised to fit with practice s on the ward , changes over time may have been seen if a defined series of time points were assessed. Regardless of these caveats, it is clear tha t the OCT washing regime does not sterilise neonatal skin, or that the microbiota is quickly reinstated with multiple organisms soon after washing. Analy sis of infa nts for whom there was both an admission swab and a weekly swab showed a similar number of colonie s were picked from both plates - this would a rgue against acquisition of OCT-tolerant organisms after initial admission. A previous study showed that for CHX after an initial decrea se in the bacterial skin burden after application, the number of recovered organisms increases and ba seline lev els are reached by approximately 72 hours [21 ]. CHX demonstrates a ‘substantive effect’, whereby the dried re sidue of . CC-BY 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted June 25, 2022. ; https://doi.org/10.1101/2022.06.23.22276813doi: medRxiv preprint 12 agent remains active in situ for a prolonged period post application onto the skin, which may reflect why a longer period is needed to repopulate the skin microbiota than after OCT exposure. The susceptibility data did not sugge st that repeated/frequent exposure to OCT selects for antiseptic tolerance on skin i sola te s (Figure 2a and 3a). In fact, isolates from BRI were significantly more susceptible to both antiseptics than those from NNUH and all isolates with highest MICs were from NNUH. This is similar to our recent comparison of the NNUH panel with a German panel (who regularly use OCT based antiseptics for skin decolonization prior to catheter insertion) and again suggests CHX exposure appea rs more likely to selec t for antiseptic tolerance than OCT[27]. The substantive effect of CHX may result in long lasting low concentrations of CHX remaining on the skin which might provide an environment for selection of tolerant mutants. Alternatively, CHX is more commonly incorpora ted in environmental cleaning wipes and products than OCT which may also reflect a greater selec tive pressure for i solate s with decreased tolerance. Strengths and L imitat i ons This is the first study to assess microbiological impacts from practising routine washing of babies with an antiseptic. This study also suggests daily OCT washing does not select for decre ased antiseptic susceptibility in CoNS and assessment of the bacterial burden of plates shows OCT washing has a limited impact on any reduction in skin microbiota. However, limitations in the study mean that possible di fferences in the genoty pes of the strains in circulation be tween the units were not assessed. The isolation periods between the sites were not exactly contemporaneous although no significant changes in units’ practice s occurred in the intervening period. Also a larg er number of isolates from NNUH were included which may skew comparison s to some degree. . CC-BY 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted June 25, 2022. ; https://doi.org/10.1101/2022.06.23.22276813doi: medRxiv preprint 13 Conc lus ion In summary, this two-site observational study shows that frequent whole-body skin washing with OCT compared to not routinely washing does not appear to result in a lasting reduction in numbers of organisms found on the skin but also does not appear to select for OCT tolerant organisms. Isolates from the NNUH were much less susceptible to antiseptics than the BRI isolates, suggesting that the bathing of neonates in the BRI NICU does not selec t for resistance. The data suggest that different antiseptic regimes can have significantly different impacts on the microbiota in terms of both compositi on and antiseptic susceptibility. Clinical tri als to systematically compare efficacy, safety and microbiological impacts of different antiseptic regimes in order to design evidence informed guidelines are lacking for this vulnerable patient group. Further work on skin cleansing in preventing neona tal sepsis is vital in order to produce best practice guideline s which will minimise infection and potential for selection of antiseptic re sistance. . CC-BY 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted June 25, 2022. ; https://doi.org/10.1101/2022.06.23.22276813doi: medRxiv preprint 14

s 1. Oranges, T., V. Dini, and M. Romanelli, S ki n Phy s i o l og y o f t he N e on at e a nd I nf an t: C l in i c al Impl icat ion s. Advances in Wound Care, 2015. 4(10): p . 587-595 .10.1089/wound.2015.0642 2. Melville, J. and T. Moss, The i mmu ne con seq ue nce s o f p r e te r m bir t h . Front. Neurosci., 2013. 7: p. 79.10.3389/fnins.2013.00079 3. Dong, Y. and C.P. Spe er, L at e- o ns et n eona t a l s ep s is : r ec e nt d eve l opm e nts . Arch. Dis. Child. Fetal Neonatal Ed., 2015. 100(3): p. F257-F263.10.1136/archdischild-2014-306213 4. Lee, J.H., Ca the ter -rela ted bloo ds tr e a m i nfec ti on s in ne on atal i nte n s iv e c ar e uni t s . Korean J. Pediatr., 2011. 54(9): p. 363 -367.10.3345/kjp.2011.54.9.363 5. Ponnusamy, V., et al., Se gme ntal p ercut a neou s c entr al v en ou s line cul ture s f or dia gn o s i s o f ca t h et er- rel ate d s ep si s . Arch. Dis. Child F etal Neonatal Ed., 2012. 97(4): p. F273- 8.10.1136/archdischild-2011-300822 6. Clarke, P., et al., S a fety an d ef ficac y o f 2% c hlorhex idine gluc o na te aq ue o us vers u s 2 % ch lorhexi dine g lucon a t e in 70 % is opr opyl alco hol f or ski n d i s in f e ctio n pr i or to percuta ne o us ce ntr a l ve n ou s cat heter i n sertio n i n pre te rm neo n ate s: th e ARCTIC ran do mise d -controlle d f e as i b il it y t r ia l p ro t o c ol. BMJ open, 2019. 9(2): p. e028022.10.1136/bmjopen-2018-028022 7. Samarasekara, S.I., D. Booth, and P. Clarke, Dev a s t atin g coag u la s e - n e ga tive s t aph y loc oc c al sep tica emi a in a n ex tr e mely low b ir th we igh t i nf an t. BMJ Case Reports, 2012. 2012.10.1136/bcr-2012-007407 8. Gupta, B., et al., E val ua tion o f e ffic acy of s k in c lean sin g w ith c hlor hexi dine i n preve nt i o n of neo n at al no soco m ial sep si s – a ra nd om ized con t roll e d tri al. J. Matern.-Fe tal Neonatal Med., 2016. 29(2): p. 242-247.10.3109/14767058.2014.996126 9. Björkqvist, M., et al., Colo niza tion pa tter n of coa gul as e -n eg ativ e stap hy loc occ i in p r e t e r m neo n ate s a nd the rel at i o n to bac tere mia. EJCMID, 2010. 29(9): p. 1085- 1093.10.1007/s10096-010-0966-3 10. Marchant, E.A., et al., Neo nat al sep si s du e t o coag ula se -ne ga tiv e stap hyloc occ i. Clin. Dev. Immunol., 2013. 2013: p. 586076.10.115 5/2013/586076 11. Ponnusamy, V., V. Venkatesh, and P. Clarke, Ski n anti se psi s in the ne on at e: w ha t sho ul d we use ? Curr. Opin. Infect. Dis., 2014. 27(3): p. 244-250.10 .1097/QCO.0000000000000064 12. Milstone, A.M., et al., Daily chl or h exid i ne bat hin g to red uc e bac tera emia i n critic ally ill ch ildren: a multice n tr e , clu s ter -ra nd omi s ed, c rosso ve r t rial . The Lancet, 2013. 381(9872): p. 1099-1106.10.1016/S0140-6736(12)61687-0 13. Montecalvo, M.A., et al., Chlorh ex idine Ba t h i ng t o Red uce Centra l Ve nou s Ca the t e r - as s oc ia ted Bl ood s t rea m In fec t io n: I mpac t an d S ustain a bili ty. Am. J. Med., 2012. 125(5): p. 505-511.10.1016/j.amjmed.2011.10.032 14. Derde, L.P.G., et al. , I nt e r v en t ions t o r educ e c o l o ni s at i o n a nd t r an sm i s s io n of an t im i cr o bi a l- r e s is ta n t b ac t e r ia i n i nt e ns i v e c a r e uni t s : a n i nt e r r upt e d t i m e s e r i es s t ud y a nd c l u s te r rand omi sed t rial . The Lanc et Infectious Diseases, 2014. 14(1): p. 31-39.10.1016/S1473- 3099(13)70295-0 15. Sankar, M.J. and V.K. Pa ul, Ef fic acy and s afe ty of whole body s k in cl ea ns ing w it h ch lorhexi dine i n neo na t e s - -a sys temic r e v iew . Pediatr I nfect Dis J, 2013. 32(6): p. e227- 34.10.1097/INF.0b013e31828693f6 16. Loveday, H.P., et a l., e pi c3: Nati on al Evi d ence -B ase d G uideli n e s fo r Pr e ve nti ng H e althcare - Assoc ia ted I nf e ction s in NHS Hos pit als i n En gla nd. J. Hosp. Infect., 2014. 86: p. S1- S70.10.1016/S0195-6701(13)60012-2 17. Dhamoda ran, M., et al., B a th i ng bab i es : c ur r e n t pr a ct i c es i n U K n e on at a l i nt e ns ive ca r e u n it s . Arch. Dis. Child Fetal Neonatal Ed., 2021. 106(2): p. 227.10.1136/archdischild-2020-320355 18. Heron, T., C. Faraday, and P. Clark e, Th e hidde n h arm s of M atchi n g Mich ig an . Arch. D is. Child. Fetal Neonatal Ed., 2013. 98(5): p. F466.10.1136/archdischild-2013-304378 . CC-BY 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted June 25, 2022. ; https://doi.org/10.1101/2022.06.23.22276813doi: medRxiv preprint 15 19. Fraser, C., et al., V aria tio n in in fe ctio n pr ev ent ion p r ac tice s for pe r i p her ally inser te d ce nt ral ve no us cat he te r s : A survey of neo na t al u nit s in Eng l an d an d Wal es . PLOS ONE, 2018. 13(11): p. e0204894.10.1371 /journal .pone.0204894 20. Sathiyamurthy, S., J. Banerjee, and S.V. Godambe, A nti s e ptic u se i n t he ne ona tal i nte n s i v e ca r e unit - a dile mma i n c linic al prac t ic e: An evid e nce ba se d revi ew. World J. Clin. Pediatr., 2016. 5(2): p. 159-171.10.5409/wjcp.v5.i 2.159 21. Johnson, J., et al., The im pac t o f chl orhex id ine gl ucona te b at hing on s k in bac teri al burde n o f neo n ate s a dmi tte d to the Neo n ata l In ten sive Care Unit . J Perinatol ., 2019. 39(1): p. 63- 71.10.1038/s41372-018-0231-7 22. Quach, C., et al., Chl orhe xidi ne ba thi ng in a terti ary c are ne ona tal in ten s iv e ca r e u nit: im p act on c entral li ne -a ssoc iate d bl o od st r eam i n fectio ns. Infect. Control Hosp. Epidemiol., 2014. 35(2): p . 158-163.10.1086/674862 23. Cleves, D., et al., Ef fect of chl orhe xidin e b ath s o n c entral -li ne- a sso ciate d blo od str e am infe cti on s in a n eon at al int en sive c are u n i t i n a de velo p ing c ou ntry. J. Hosp. Infect., 2018. 100(3): p. e196-e199.10.1016/j.jhin.2018.03.022 24. Baier, C., et al ., Uni ver sal de col on izatio n with oc teni din e : Firs t expe r i ence s i n a t er tiary burn inten si v e ca r e uni t. Burns Open, 2019. 3(1): p. 8-11.10.1016/j.burnso.2018.10.003 25. Gastmeier, P., et al., An ob s erv at i o nal st u dy of t he univ ersal u s e o f octe ni dine t o d ecrea s e no soco mial bl oo d s tre am in fec t io n s a n d MDR or gani s ms. J Antimicrob Chemothe r., 2016. 71(9): p . 2569-2576.10.1093/jac/dkw170 26. Messler, S., et al., R edu cti on o f n osoc omi al bloo d str e am in fec tio ns an d no s oc omi a l va ncomy cin-re si sta n t Enter ococ cus faec i um o n an in te n s iv e c ar e unit af ter in trod uc tion o f anti se ptic octen i dine -b a sed b ath in g. J. Hosp. Infect., 2019. 101(3): p. 264- 271.10.1016/j.jhin.2018.10.023 27. Sethi, D.K., et al ., Chl orhe xidin e gluc on at e us age i s a ssoci at e d wi t h a nti s e pt i c tol e ran ce i n sta phyloc occ i from t he ne ona tal i nte nsi v e c are uni t. J Antimicrob Chemother., 2021. 3(4).10.1093/jacamr/dlab173 28. EUCAST, EUCA S T . 2021 29. Sethi, D.K., et al ., Chl orhe xidin e gluc on at e us age i s a ssoci ate d wi t h a nti s e ptic tol e r an ce i n sta phyloc occ i from t he ne ona tal i nte nsi v e c are uni t. JAC Antimicrob Resist, 2021. 3(4): p. dlab173.10.1093/jacamr/dlab173 30. Htun, H.L., et al., Ch l orhexi di ne an d octe n idine use , carriage of qac gene s, an d re duce d anti se ptic susce p tibili ty in met hi cilli n-r e s i st an t Stap hy loco ccu s au r e u s i s ol a t e s fr o m a hea lth care ne t w ork. Clin. Microbiol. Infect., 2019. 25(9): p. 1154.e1- 1154.e7. https://doi.org/10.1016/j.cmi.2018.12.036 31. Younge, N.E., et al., E ar l y-li fe skin microb i ota i n ho spi tali zed pr e t e r m a nd full - t erm i nfa nt s . Microbiome, 2018. 6.10.1186/s40168-018-0486-4 . CC-BY 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted June 25, 2022. ; https://doi.org/10.1101/2022.06.23.22276813doi: medRxiv preprint 16 Fig ur e s Fig ur e 1: Susceptibility of isolates from BRI to OCT and CHX showed no correlation (p= 0.4, according to Spearman test). Fig ur e 2: A Comparison of MICs of OCT against isolates from BRI where regular daily whole-body OCT washing was in place (n=78) and isolate s from NNUH where there wa s no regular washing of neona te s while in NICU (n=863). B Box plot showing numbers of isolates with different OCT MICs from each site (** ** P = <0.0001). Thin horizontal line indicates the mean and whiskers standard error. Fig ur e 3: A Comparison of MICs of CHX against isolates BRI where regular da ily whole-body OCT washing was in place (n=78) and isolates from NNUH where there wa s no regular washing of neona te s while in NICU (n=863). B Box plot showing numbers of isol ates with different CHX MICs from each site (** ** P = <0.0001). Thin horizontal line indicates the mean and whiskers standard error. . CC-BY 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted June 25, 2022. ; https://doi.org/10.1101/2022.06.23.22276813doi: medRxiv preprint 17 Ack nowledge ment s We sinc erely thank all NNUH and BRI research and clinical nurses who helped with skin swa b collection. We are gra te ful to Julie Dawson, Research Services Manager at NNUH for reviewing our st udy protocol. Conflic t of int e r est None Sourc e of Funding This work was supported by an award from the Biotechnology and Biological Sciences Research Council (BB/T014644/1 ). . CC-BY 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted June 25, 2022. ; https://doi.org/10.1101/2022.06.23.22276813doi: medRxiv preprint OCT MIC mg/ml % isolates 0.125 0.250 0.500 1.000 2.000 4.000 8.000 16.00032.00064.000 0 20 40 60 80 Bradford NNUH OCT MIC mg/ml Bradford NNUH 0.0625 0.125 0.25 0.5 1 2 4 8 16 32 64 A B . CC-BY 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted June 25, 2022. ; https://doi.org/10.1101/2022.06.23.22276813doi: medRxiv preprint CHX MIC mg/ml % isolates 0.125 0.250 0.500 1.000 2.000 4.000 8.000 16.00032.00064.000 0 20 40 60 Bradford NNUH CHX MIC mg/ml Bradford NNUH 0.03125 0.125 0.5 2 8 32 128 A B . CC-BY 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted June 25, 2022. ; https://doi.org/10.1101/2022.06.23.22276813doi: medRxiv preprint OCT MIC mg/ml CHX MIC mg/ml 0.0 0.5 1.0 1.5 0.0 0.5 1.0 1.5 . CC-BY 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted June 25, 2022. ; https://doi.org/10.1101/2022.06.23.22276813doi: medRxiv preprint