Catheter-Related Bloodstream Infection in Patients Receiving Vitamin-Enriched Peripheral Parenteral Nutrition: A Retrospective Cohort Study

Abstract

In vitro studies have reported that multiple vitamins may promote bacterial growth and infection. In clinical practice, patients receiving peripheral parenteral nutrition (PPN) may develop catheter-related bloodstream infections (CRBSIs). Since some PPN formulations contain multiple vitamins, they may increase the risk of CRBSIs. Therefore, the present study investigated the relationship between PPN infusions containing vitamins and the incidence of CRBSIs and examined the effects of different antiseptics used before catheter insertion. Patients were divided into the following groups: those receiving the BFLUID injection (1 vitamin), those receiving the PAREPLUS injection (9 vitamins), those disinfected with povidone iodine before the PAREPLUS injection, and those disinfected with chlorhexidine alcohol before the PAREPLUS injection. We analyzed infection rates in each group. The PAREPLUS group showed a significantly higher rate of infections by Staphylococcus species. Furthermore, a longer duration of PPN administration (≥2 weeks) was associated with a higher risk of CRBSI across all detected pathogens. Vitamin-enriched PPN increased approximately 2-fold in CRBSI compared with a thiamine-only formulation, independent of infusion duration. However, the incidence of CRBSI did not significantly differ between the different antiseptic groups. These results suggest that proper infection control and careful management are essential during PPN therapy, especially when multiple vitamin products are used over extended periods.

INTRODUCTION

Nutritional management plays a crucial role in both therapeutic interventions and health maintenance. The absence of appropriate nutritional support in malnourished patients leads to a significant decline in survival rates.¹⁾ In clinical practice, 2 primary methods of nutritional support are employed: enteral nutrition and parenteral nutrition. Their selection is based on the patient’s clinical condition; however, parenteral nutrition is generally selected when enteral nutrition is not feasible. Parenteral nutrition is further classified into peripheral parenteral nutrition (PPN) and total parenteral nutrition (TPN). PPN is typically administered when nutritional support is required for less than 2 weeks, whereas TPN is indicated for cases requiring comprehensive nutritional supplementation over periods exceeding 2 weeks. PPN may also be used for durations longer than 2 weeks depending on the patient’s condition.

PPN infusions contain glucose, electrolytes, amino acids, and other components, and offer several advantages over TPN infusions, including a simpler procedure, easier management, and a lower risk of complications associated with catheter insertion and maintenance. During TPN, vitamin B₁ deficiency can easily occur, potentially leading to serious conditions such as Wernicke’s encephalopathy and lactic acidosis.^2–6) PPN is also associated with a risk of vitamin B₁ deficiency. To mitigate this issue, preparations such as the BFLUID injection containing vitamin B₁ have been developed and are now widely used in clinical practice. Previous studies have demonstrated that blood concentrations of hydrophilic vitamins generally decrease before and after gastrointestinal surgery.⁷⁾ The PAREPLUS injection, containing 9 types of hydrophilic vitamins, has also been introduced for clinical use. The addition of vitamins in parenteral nutrition is recommended by the American Society for Parenteral and Enteral Nutrition.⁸⁾ However, these conditions may provide a favorable environment for bacterial proliferation. Catheter-related bloodstream infections (CRBSIs) are often caused by the commensal skin flora. The growth of these bacteria is promoted by hydrophilic vitamins. The PAREPLUS injection, which contains 9 types of vitamins, has been shown to support the proliferation of the skin flora more than other infusion formulations.⁹⁾ Table 1 shows the compositions of the BFLUID and PAREPLUS injections based on the official package inserts of products available in Japan.

Table 1. Components of BFLUID and PAREPLUS Injections

	BFLUID injection	PAREPLUS injection
Volume	500 mL	500 mL
Calories	210 kcal	210 kcal
pH	Approximately 6.7	Approximately 6.9
Osmotic pressure ratio (ratio to physiological saline)	Approximately 3	Approximately 3
Nonprotein calories/nitrogen ratio	64	64
Carbohydrate
Glucose	37.50 g	37.499 g
Amino acids
Total free amino acid content	15.00 g	15.00 g
Branched-chain amino acid composition	30%	30%
Essential amino acids/nonessential amino acids	1.79	1.79
Total nitrogen content	2.35 g	2.35 g
Electrolytes
Na+	17.5 mEq	17.1 mEq
K+	10 mEq	10 mEq
Mg2+	2.5 mEq	2.5 mEq
Ca2+	2.5 mEq	2.5 mEq
Cl−	17.5 mEq	17.6 mEq
SO42−	2.5 mEq	2.5 mEq
Acetate−	8 mEq	0.6 mEq
Lactate−	10 mEq	12.7 mEq
Citrate3−	3 mEq	6 mEq
P	5 mmol	5 mmol
Zn	2.5 µmol	2.4 µmol
Vitamins
Thiamine chloride hydrochloride	0.96 mg	1.91 mg
Riboflavin 5′-phosphate sodium	—	1.25 mg
Pyridoxine hydrochloride	—	1.25 mg
Cyanocobalamin	—	2.5 µg
Ascorbic acid	—	50 mg
Nicotinamide	—	10 mg
Panthenol	—	3.75 mg
Biotin	—	25 µg
Folic acid	—	0.1 mg

Among bacteria in the skin flora, Staphylococcus aureus proliferates well in the presence of nicotinic acid or nicotinamide.^9,10) Therefore, the PAREPLUS injection, which contains 9 vitamins including nicotinamide, is associated with a higher risk of infection by methicillin-resistant S. aureus (MRSA) than the BFLUID injection, which contains only thiamine. Additionally, biotin has been reported to promote the growth of Candida species.^11,12)

CRBSIs are associated with high mortality rates in patients receiving hemodialysis.¹³⁾ According to the guidelines issued by the Infectious Diseases Society of America, treatment durations for MRSA-related CRBSIs are longer than those for infections caused by other pathogens, underscoring the importance of preventing nosocomial MRSA infections.^14–17)

CRBSI remains a substantial source of morbidity, mortality, and healthcare costs, with pooled incidence rates of 1.5–2.5 per 1000 catheter-days in mixed medical–surgical wards.¹⁷⁾ Although central venous catheters account for the majority of episodes, peripheral venous catheters are increasingly recognized as important contributors to nosocomial bacteremia.¹⁸⁾ PPN is routinely prescribed when short-term parenteral support (<14 d) is required. PPN bags are formulated with neutral pH and low osmolality to limit phlebitis, yet these physicochemical properties, compounded by nutrient richness, may favor microbial growth.

Hydrophilic vitamins, particularly nicotinamide and biotin, enhance S. aureus and Candida spp. proliferation in vitro.⁹⁾ In Japan, 2 licensed PPN products dominate adult practice: BFLUID^® (thiamine only) and PAREPLUS^® (thiamine plus 8 additional hydrophilic vitamins). Whether differences in vitamin composition translate into divergent CRBSI rates has not been evaluated clinically. Concurrently, guidelines recommend chlorhexidine alcohol over povidone iodine for skin antisepsis during peripheral catheter insertion, yet real-world effect sizes in PPN cohorts remain inconsistent.

We therefore performed a retrospective cohort study to (i) quantify the impact of vitamin-enriched PPN (PAREPLUS) versus thiamine-only PPN (BFLUID) on CRBSI incidence, and (ii) explore whether switching the insertion-site antiseptic from povidone-iodine to chlorhexidine-alcohol modified infection risk.

MATERIALS AND METHODS

Selection of Subjects

Subjects were inpatients from all departments at the Teikyo University Chiba Medical Center who received the BFLUID injection between April 2013 and March 2014 (antiseptic: 10% povidone iodine) and those who received the PAREPLUS injection between April 2015 and March 2016 (antiseptic: 10% povidone iodine).

We switched from BFLUID infusion to PAREPLUS infusion in April 2015 and have not used BFLUID infusion in our hospital. Therefore, data collection was conducted for 1 year, starting from April 2015. In fiscal year 2014, both BFLUID and PAREPLUS were used during the transition period; therefore, the data for BFLUID from 2013 was used instead. Our institution transitioned from povidone-iodine to chlorhexidine-alcohol preparations for disinfection because the latter is more effective at preventing colony formation during peripheral venous catheter insertion.^19–21) Accordingly, patients who received PAREPLUS injection (disinfectant: 1% chlorhexidine alcohol) between April 2023 and March 2024, after the complete transition from 10% povidone iodine to 1% chlorhexidine alcohol, were also included in this study.

Exclusion Criteria

Patients who received the BFLUID injection mixed with various vitamin preparations were excluded.

Catheter Insertion and Maintenance Protocol

Peripheral catheters were inserted by a physician or nurse using the institution’s standard precautions (hand hygiene, clean gloves, and transparent dressings). Maximum barrier precautions are not required for peripheral lines.

Determination of CRBSI

Patients who were determined by our hospital’s infection control department to have contamination not requiring treatment were classified as not having CRBSI.

Statistical Analysis

Continuous data are presented as medians and compared with the Mann–Whitney U test. Categorical variables are presented as number (%) and compared with the chi-squared test or Fisher’s exact test. Multivariable logistic regression estimated adjusted odds ratios for CRBSI. Statistical significance was defined as two-sided p < 0.05. All statistical analyses were performed using statistical software (StatFlex Ver. 7; Artec Co., Ltd., Osaka, Japan).

Ethical Considerations

The present study was conducted under the ethical principles outlined in the “Declaration of Helsinki” and the “Ethical Guidelines for Life Science and Medical Research Involving Human Subjects.” Since this was a retrospective study, it falls under the category of “research using existing specimens and information held at one’s own research institution,” as defined in the Ethical Guidelines for Life Science and Medical Research Involving Human Subjects (revised April 1, 2022). Therefore, with approval from the Ethics Committee, individual informed consent for the use of clinical information from each patient was not obtained; an opt-out approach was instead adopted. An information disclosure document outlining the research title, objectives, significance, subjects, methods, affiliated institution, handling of personal information, and contact information was posted within the hospital and was also made available on the Ethics Committee’s website. This ensured that potential subjects had the opportunity to decline inclusion. The present study was conducted after obtaining approval from the Teikyo University Ethical Review Board for Medical and Health Research Involving Human Subjects (Approval No. 24-080).

RESULTS AND DISCUSSION

As described in the selection criteria, the following data were collected from the medical records of all subjects: age, sex, duration of the PPN injection, and blood culture results. Table 2 shows the baseline characteristics of subjects (sex, age [all subjects], age [>65 years], and duration of the PPN injection with the median or ratio). The results obtained showed no significant differences in any characteristics between the BFLUID and PAREPLUS injection groups, indicating no bias between the 2 groups.

Table 2. Baseline Characteristics of Subjects

	BFLUID injection (N = 539)	PAREPLUS injection (N = 552)	p-Value
Age, median (IQR) (years)	73.0 (64.3–79.0)	72.0 (62.0–80.0)	0.61
Age, >65 years, No. (%)	404 (75.0)	389 (70.5)	0.10
Sex (male/female), No.	318/221	326/226	0.98
Duration of the PPN injection, median (range) (weeks)	0.70 (0.40–1.40)	0.70 (0.30–1.30)	0.29

Table 3 shows the results of the analysis of infection-related factors in patients who underwent blood cultures, including all detected bacteria. Among these patients, the infection rate was significantly higher in the PAREPLUS injection group than in the BFLUID injection group. No significant differences were observed in sex, overall age, or age >65 years. Additionally, a longer duration of PPN infusion correlated with an increased infection rate. Accordingly, the administration of PPN for less than 2 weeks is recommended. If unavoidable circumstances necessitate longer use, clinicians need to be aware of the significantly increased risk of CRBSI. A previous study reported that the appropriate use of antimicrobial agents in the early stages of bloodstream infections promoted better outcomes, significantly reducing mortality and adverse events in patients with bloodstream infections, while also improving the rate of de-escalation.²²⁾ Since CRBSI increases medical costs and the length of hospitalization,^15,16) adhering to established guidelines for the duration of PPN and ensuring the proper disinfection of the catheter insertion site or hub may reduce the incidence of CRBSI, healthcare costs, and hospitalization periods.²³⁾

Table 3. Factors Contributing to Bloodstream Infection (Bacteria Detected from Blood Cultured)

*PPN type: BFLUID injection and PAREPLUS injection.

Table 4 shows the types and number of bacteria identified in the present study. During the entire study period, 12 different microbial species were detected, most of which were Staphylococcus species and Candida species. Thus, Staphylococcus species and Candida species were detected across all groups, and an analysis of infection-related factors was performed specifically for these organisms.

Table 4. Types and Number of Bacteria Identified in the Present Study

Bacteria type	BFLUID injection (10% povidone iodine), N = 539	PAREPLUS injection (10% povidone iodine), N = 314	PAREPLUS injection (1.0% chlorhexidine alcohol), N = 238
Bacillus species	1	0	0
Bacteroides species	0	1	0
Citrobacter species	1	0	0
Clostridium species	0	1	0
Enterobacter species	2	0	0
Enterococcus species	1	0	0
Escherichia species	1	0	1
Klebsiella species	1	0	1
Pseudomonas species	4	0	0
Serratia species	1	0	0
Staphylococcus species	11	15	17
Candida species	1	1	5
Total	24	18	24

Table 5 shows the results of the analysis of infection-related factors for Staphylococcus species. A significant difference was observed between the BFLUID and PAREPLUS injection groups, with the use of the PAREPLUS injection being associated with a significantly higher incidence of infection. These results suggest that administering PPN with a high vitamin content may increase the risk of CRBSI. Therefore, when administering multiple vitamin PPN, it is essential to implement thorough infection control measures to prevent CRBSI.

Table 5. Factors Contributing to Staphylococcus Species Infection

*PPN type: BFLUID injection and PAREPLUS injection.

Previous in vitro studies reported that within Staphylococcus species, the response to the presence or absence of vitamins may vary depending on the specific species or strain.⁹⁾ In the present study, analyses were conducted collectively for Staphylococcus species. However, future investigations focusing on specific species or strains may provide more detailed insights into the mechanisms underlying CRBSI.

No significant differences were noted in sex, overall age, or age >65 years. On the other hand, the duration of PPN administration correlated with infection, suggesting that a prolonged infusion increased the risk of infection by Staphylococcus species.

Among Staphylococcus species, methicillin-resistant coagulase-negative staphylococci have been reported to account for more than 30% of CRBSI cases.^13,24) Accordingly, the increased risk of infections must be carefully considered when administering vitamin-containing PPN.

Table 6 shows the results of the analysis of infection-related factors for Candida species. In the present study, no significant difference was observed in the incidence of Candida species infections between the BFLUID and the PAREPLUS injection groups. In vitro studies previously reported that Candida species, including strains IFM40009 and IFM61197, proliferated even in the absence of vitamins and that their growth was significantly promoted in multiple vitamin environments.⁹⁾ Therefore, no significant difference was noted in Candida species between the BFLUID and PAREPLUS injection groups. On the other hand, the presence of biotin has been suggested to increase the risk of infection by Candida species. Since antiseptics were not used in in vitro experiments, the differences observed between in vitro and clinical findings may be attributed to the presence or absence of antiseptics.^9,11,12)

Table 6. Factors Contributing to Candida Species Infection

*PPN type: BFLUID injection and PAREPLUS injection

The results obtained also showed that the duration of the PPN infusion correlated with the incidence of infection. No significant differences were observed in sex, overall age, or age >65 years.

Based on the results presented in Tables 5 and 6, longer durations of PPN administration correlated with an increased incidence of infection by all bacterial types. Table 7 shows the relationship between infusion duration and infection rates. Among all patients in the BFLUID and PAREPLUS injection groups, 934 received the PPN infusion for ≤2 weeks, while 157 received it for >2 weeks. In total, 32 of the 934 patients treated for ≤2 weeks and 33 of the 157 patients treated for >2 weeks developed infections. Statistical analyses using the Mann–Whitney U test and Fisher’s exact test revealed that patients who received the PPN infusion for more than 2 weeks were significantly more likely to develop infections (p < 0.01). These results suggest that for both types of PPN infusions, the risk of infection markedly increased when the duration of administration exceeded 2 weeks.

Table 7. Relationships between Infusion Durations and Infection Rates

Infusion duration (weeks)	BFLUID injection (10% povidone iodine)	PAREPLUS injection (10% povidone iodine)	PAREPLUS injection (1.0% chlorhexidine alcohol)
	Number of infections/patients (ratio; %)
0–0.9	3/313 (0.96)	2/187 (1.07)	5/153 (3.27)
1.0–1.9	7/135 (5.19)	7/80 (8.75)	5/49 (10.20)
2.0–2.9	6/45 (13.33)	0/25 (0)	8/23 (34.78)
3.0–3.9	1/22 (4.55)	3/8 (37.50)	1/7 (14.29)
4.0–4.9	1/7 (14.29)	2/7 (28.57)	2/2 (100)
5.0–5.9	2/9 (22.22)	1/2 (50.00)	0/1 (0)
6.0–6.9	1/4 (25.00)	3/4 (75.00)	0/0 (0)
7.0–7.9	1/1 (100)	0/0 (0)	1/1 (100)
8.0–8.9	0/1 (0)	0/0 (0)	0/0 (0)
9.0–9.9	1/1 (100)	0/0 (0)	0/0 (0)
10.0–10.9	0/1 (0)	0/0 (0)	0/0 (0)
11.0–11.9	0/0 (0)	0/0 (0)	0/0 (0)
12.0–12.9	0/0 (0)	0/1 (0)	1/1 (100)
13.0–13.9	0/0 (0)	0/0 (0)	1/1 (100)
Total	23/539 (4.27)	18/314 (5.73)	24/238 (10.08)

According to Japanese clinical guidelines, PPN infusions are recommended for use within 2 weeks for parenteral nutrition. However, administration may be extended beyond this period in several cases depending on the patient’s condition. Due to the increased risk of bacterial infection associated with prolonged use, adherence to the guidelines limiting PPN infusion to within 2 weeks is considered desirable. In addition, since the effect of disinfection is not long-lasting, frequent cleaning and disinfection of the insertion site may be necessary to prevent bacterial infections.

Table 8 shows the effects of pre-catheter insertion antiseptics on the incidence of infection during the PAREPLUS injection. Regarding the impact of disinfection with 10% povidone iodine or 1.0% chlorhexidine alcohol on infection rates, no significant differences were observed for any bacterial or fungal genus. Although previous studies reported that chlorhexidine exhibited stronger antimicrobial activity than povidone-iodine formulations,^19–21) no significant differences were observed in the present study. In previous studies (references^19–21)), the authors either did not specify the concentration of chlorhexidine, used 2% chlorhexidine, or employed a mixed solution containing other disinfectants such as benzalkonium chloride. Since this study used 1% chlorhexidine alcohol, it was considered necessary to either increase the concentration of chlorhexidine alcohol or use a mixed solution with other disinfectants in the future.

Table 8. Antiseptic Solution before Catheterization

	10% Povidone iodine (N = 314)	1.0% Chlorhexidine alcohol (N = 238)	p-Value
Bacteria detected from blood cultures, No. (%)
Staphylococcus species	15 (4.8)	17 (7.1)	0.24
Candida species	1 (0.3)	5 (2.1)	0.09
Other bacteria	2 (0.6)	2 (0.8)	1.00

The chi-squared test and Fisher’s exact test were used for group comparisons.

The present study was conducted at a single institution and included only a limited number of risk factors. The catheter infection rate at our hospital tended to be slightly higher than the average reported by JANIS (Japan Nosocomial Infections Surveillance). On the other hand, the infection rate at our hospital was lower than the catheter-related infection rates reported in studies conducted in mixed medical–surgical wards.

Therefore, it is considered that future evaluations involving multiple institutions and a greater variety of risk factors will enable a more detailed assessment of the relationship between vitamin preparations and CRBSI. In addition, analyzing the causes of catheter-related infections not only from PPN infusions but also from all types of catheter placements may help reduce the catheter infection rate.

CONCLUSION

This study provides the first clinical evidence that vitamin-enriched PPN is associated with an approximately two-fold increase in CRBSI compared with a thiamine-only formulation, independent of infusion duration. The finding accords with mechanistic in vitro research demonstrating that nicotinamide and other hydrophilic vitamins potentiate Staphylococcus growth. Our data also emphasize infusion duration as a modifiable risk factor; adherence to the recommended ≤14-d course halved infection risk.

The future identification of individual strains detected in clinical settings, with further data accumulation and analyses, will contribute to a more detailed understanding of the mechanisms underlying CRBSI associated with various PPN formulations.

Conflict of Interest

The authors declare no conflict of interest.

REFERENCES

• 1)  Kagansky N, Berner Y, Koren-Morag N, Perelman L, Knobler H, Levy S. Poor nutritional habits are predictors poor outcome in very old hospitalized patients. Am. J. Clin. Nutr., 82, 784–791, quiz, 913–914 (2005).
• 2)  Nadel AM, Burger PC. Wernicke encephalopathy following prolonged intravenous therapy. JAMA, 235, 2403–2405 (1976).
• 3)  Velez RJ, Myers B, Guber MS. Severe acute meta bolic acidosis (acute beriberi): an avoidable complication of total parenteral nutrition. JPEN J. Parenter. Enteral Nutr., 9, 216–219 (1985).
• 4)  Abdalian R, Fernandes G, Duerksen D, Jeejeebhoy KN, Whittaker S, Gramlich L, Allard JP. Prescription of trace elements in adults on home parenteral nutrition: current practice based on the Canadian Home Parenteral Nutrition Registry. JPEN J. Parenter. Enteral Nutr., 37, 410–415 (2013).
• 5)  Jin J, Mulesa L, Carrilero Rouillet M. Trace elements in parenteral nutrition: considerations for the prescribing clinician. Nutrients, 9, 440 (2017).
• 6)  Fedeli P, Davies RJ, Cirocchi R, Popivanov G, Bruzzone P, Giustozzi M. Total parenteral nutrition-induced Wernicke’s encephalopathy after oncologic gastrointestinal surgery. Open Med., 15, 709–713 (2020).
• 7)  Ozasa H, Ishibashi N, Ikeda S, Imaizumi T, Miyagi M, Yano S, Aoyagi K, Akagi Y, Ogata Y, Shirouzu K. Clinical examination of the water-soluble vitamin levels in blood during peripheral parenteral nutrition. Kurume Med. J., 53, 79–87 (2006).
• 8)  Strausburg KM. Parenteral nutritional admixture. Chapter 8, The ASPEN Nutrition Support Practice Manual. American Society for Parenteral and Enteral Nutrition, Silver Spring, pp. (8-1)–(8-3) (1998).
• 9)  Omotani S, Tani K, Nagai K, Hatsuda Y, Mukai J, Myotoku M. Water soluble vitamins enhance the growth of microorganisms in peripheral parenteral nutrition solutions. Int. J. Med. Sci., 14, 1213–1219 (2017).
• 10)  Stokes JL, Gunness M, Foster JW. Vitamin content of ingredients of microbiological culture media. J. Bacteriol., 47, 293–299 (1944).
• 11)  Firestone BY, Koser SA. Growth promoting effect of some biotin analogues for Candida albicans. J. Bacteriol., 79, 674–676 (1960).
• 12)  Hasim S, Tati S, Madayiputhiya N, Nandakumar R, Nickerson KW. Histone biotinylation in Candida albicans. FEMS Yeast Res., 13, 529–539 (2013).
• 13)  Wisplinghoff H, Bischoff T, Tallent SM, Seifert H, Wenzel RP, Edmond MB. Nosocomial bloodstream infections in U.S. hospitals: analysis of 24,179 cases from a prospective nationwide surveillance study. Clin. Infect. Dis., 39, 309–317 (2004).
• 14)  Dimick JB, Pelz RK, Consunji R, Swoboda SM, Hendrix CW, Lipsett PA. Increased resource use associated with catheter-related bloodstream infection in the surgical intensive care unit. Arch. Surg., 136, 229–234 (2001).
• 15)  Blot SI, Depuydt P, Annemans L, Benoit D, Hoste E, De Waele JJ, Decruyenaere J, Vogelaers D, Colardyn F, Vandewoude HK. Clinical and economic outcomes in critically ill patients with nosocomial catheter-related bloodstream infections. Clin. Infect. Dis., 41, 1591–1598 (2005).
• 16)  Warren DK, Quadir WW, Hollenbeak CS, Elward AM, Cox MJ, Fraser VJ. Attributable cost of catheter-associated bloodstream infections among intensive care patients in a nonteaching hospital. Crit. Care Med., 34, 2084–2089 (2006).
• 17)  Mermel LA, Allon M, Bouza E, Craven DE, Flynn PF, O’Grady P, Raad II, Rijnders BJA, Sherertz RJ, Warren DK. Clinical practice guidelines for the diagnosis and management of intravascular catheter-related infection: 2009 update by the Infectious Diseases Society of America. Clin. Infect. Dis., 49, 1–45 (2009).
• 18)  Marsh N, Larsen EN, Takashima M, Kleidon T, Keogh S, Ullman AJ, Mihala G, Chopra V, Rickard CM. Peripheral intravenous catheter failure: a secondary analysis of risks from 11,830 catheters. Int. J. Nurs. Stud., 124, 104095 (2021).
• 19)  Chaiyakunapruk N, Veenstra DL, Lipsky BA, Saint S. Chlorhexidine compared with povidone-iodine solution for vascular catheter-site care: a meta-analysis. Ann. Intern. Med., 136, 792–801 (2002).
• 20)  Guenezan J, Marjanovic N, Drugeon B, Neill RO, Liuu E, Roblot F, Palazzo P, Bironneau V, Prevost F, Paul J, Pichon M, Boisson M, Frasca D, Mimoz O. Chlorhexidine plus alcohol versus povidone iodine plus alcohol, combined or not with innovative devices, for prevention of short-term peripheral venous catheter infection and failure (CLEAN 3 study): an investigator-initiated, open-label, single centre, randomised-controlled, two-by-two factorial trial. Lancet Infect. Dis., 21, 1038–1048 (2021).
• 21)  Mimoz O, Villeminey S, Ragot S, Dahyot-Fizelier C, Laksiri L, Petitpas F, Debaene B. Chlorhexidine-based antiseptic solution vs. alcohol-based povidone-iodine for central venous catheter care. Arch. Intern. Med., 167, 2066–2072 (2007).
• 22)  Niwa T, Watanabe T, Goto T, Ohta H, Nakayama A, Suzuki K, Shinoda Y, Tsuchiya M, Yasuda K, Murakami N, Itoh Y. Daily review of antimicrobial use facilitates the early optimization of antimicrobial therapy and improves clinical outcomes of patients with bloodstream infections. Biol. Pharm. Bull., 39, 721–727 (2016).
• 23)  Maki DG, Stolz SM, Wheeler S, Mermel LA. Prevention of central venous catheter-related bloodstream infection by use of an antiseptic-impregnated catheter: a randomized, controlled trial. Ann. Intern. Med., 127, 257–266 (1997).
• 24)  Richards MJ, Edwards JR, Culver DH, Gaynes RP. Nosocomial infections in medical intensive care units in the United States. Crit. Care Med., 27, 887–892 (1999).