Randomized Controlled Trial of Antiseptic Hand Hygiene Methods in an Outpatient Surgery Clinic

Introduction. Outpatient wound care plays an integral part in any plastic surgery practice. However, compliance with hand hygiene measures has shown to be low, due to skin irritation and lack of time. The objective of this trial was to determine whether single-use, long-acting antiseptics can be as effective as standard multiple-use hand hygiene methods in an outpatient surgical setting. Materials and Methods. A prospective, randomized controlled trial was performed in the authors’ outpatient plastic surgery clinic at Rutgers New Jersey Medical School, Newark, NJ to compare the efficacy of an ethyl alcohol-based sanitizer (Avagard D Instant Hand Aniseptic, 3M Health Care, St. Paul, MN), a benzalkonium chloride-based sanitizer (Soft & Shield, Bioderm Technologies, Inc, Trenton, NJ, distributed by NAPP Technologies, Hackensack, NJ ), and soap-and-water handwashing. Subjects included clinic personnel, who were followed throughout the course of a 3-hour clinic session with hourly hand bacterial counts taken. Results. During the course of the trial, 95 subjects completed the clinic session utilizing 1 of the hand hygiene methods (36 ethyl alcohol-based sanitizer, 38 benzalkonium chloride-based sanitizer, and 21 soap-and-water handwashing). There was no difference between hand bacterial counts using the different methods at 4 hourly time points (P > 0.05). Hand bacterial counts increased significantly over the 3-hour clinic session with the ethyl alcohol-based sanitizer (9.24 to 21.90 CFU, P < 0.05), benzalkonium chloride-based sanitizer (6.69 to 21.59 CFU, P < 0.05), and soap-and-water handwashing (8.43 to 22.75 CFU, P < 0.05). Conclusion. There does not appear to be any difference in efficacy between single-use, long-acting sanitizer, and standard multiple-use hand hygiene methods. Hand bacterial counts increased significantly over the course of the 3-hour clinic session regardless of the hand hygiene measure used. Hand condition of subjects was improved with the ethyl alcohol-based sanitizer and the benzalkonium chloride-based sanitizer compared with soap-and-water handwashing.

With the extensive literature documenting a lack of compliance with hand hygiene methods in clinical settings, drastic measures have been taken by hospital systems to improve compliance and control hospital-acquired infections. Such measures include implementation of complex technological systems^1-3 or direct monitoring of hand hygiene.⁴ It is unrealistic, however, to implement these hand hygiene systems in a broad-based manner considering the necessary costs and manpower.

There are various reasons for the lack of compliance with hand hygiene measures, including skin irritation, inaccessibility to handwashing supplies, inconvenient placement of sinks, neglect, high workloads, and lack of time.^5-8 Therefore, it is not surprising that the preference for a particular hand hygiene agent by health care workers is determined by its antiseptic efficacy, cost, smell, consistency, color, tendency to cause skin irritation, and time needed for drying.⁹

Popular antimicrobial agents include those based on alcohol, chlorhexidine, chlorine, hexachlorophene, iodine, chloroxylenol, quaternary ammonium compounds, and triclosan.¹⁰ While alcohol-based hand antiseptics have been found to rapidly reduce hand bacteria counts more than soaps or detergents containing hexachlorophene, povidone-iodine, 4% chlorhexidine, or triclosan,^11,12 they perform poorly against bacterial spores, irritate the skin, provide no barrier protection, and have little long-lasting activity.⁹

Soft & Shield (NAPP Technologies, Hackensack, NJ) is an alcohol-free, hypoallergenic hand sanitizer and moisturizer. The active agent in the sanitizer is 0.13% benzalkonium chloride. The sanitizer works via dual methods by forming a protective barrier on the skin and via long-lasting antimicrobial activity for up to 4 hours with only 1 application.

This study specifically compares the in vivo efficacy of 3 different hand hygiene methods in their ability to remove resident flora from the hands of medical personnel. The products compared in this trial were Avagard D (3M Health Care, St. Paul, MN), an alcohol-based (61% ethanol) sanitizer used throughout the University Hospital in Newark, NJ, a benzalkonium chloride-based sanitizer, and soap-and-water handwashing.

This study was a prospective, randomized equivalence trial comparing the antiseptic efficacy of an alcohol-based sanitizer, a benzalkonium chloride-based sanitizer, and soap-and-water handwashing in their respective ability to remove resident flora from the hands of medical personnel. Eligible subjects included full-time employees, defined as individuals who worked > 30 hours per week) in the Division of Plastic and Reconstructive Surgery of Rutgers New Jersey Medical School, Newark, NJ, who were between the ages of 18-65 years. Exclusion criteria included individuals with an allergy to the  study products or to latex, who were receiving topical or systematic steroids or antibiotics, and those with diagnosed dermatologic conditions such as psoriasis.

The study protocol was reviewed and approved by the Institutional Review Board representing Rutgers New Jersey Medical School. Subjects who agreed to volunteer for the study and met criteria were randomly assigned to 1 of the 3 hand hygiene regimens. Upon enrollment in the study, each subject completed a demographic survey. Each trial lasted 3 hours at the authors’ outpatient clinic. The participants began by removing artificial nails, nail polish, and hand jewelry. Subsequently, all subjects washed their hands with nonantimicrobial liquid soap using the handwashing protocol described in this paper before being given the alcohol-based sanitizer or the benzalkonium chloride-based sanitizer. Subjects in the soap-and-water handwashing group were not provided an additional antiseptic product. Subjects subsequently treated patients (except those with confirmed Clostridium difficile or multidrug-resistant infections) during the division’s outpatient clinic for a 3-hour period. Subjects were instructed to wear gloves whenever contacting patients.

For those in the benzalkonium chloride-based sanitizer group, 0.5 mL of product was dispensed into the subject’s nondominant palm, and then rubbed into the fingers and palm until well dispersed. Hands were rinsed with cold water during the trial only if visibly soiled. The product was not reapplied throughout the course of the clinic.

For those in the ethyl alcohol-based sanitizer group, 2-3 mL of product was dispensed into the subject’s nondominant palm, and then rubbed into the fingers and palm until well dispersed. The product was reapplied prior to each new patient contact.

Subjects in the soap-and-water handwashing group were instructed to wash their hands between every patient encounter. For each washing, subjects were instructed to rinse two-thirds of their forearms with tap water for 30 seconds, scrub hands with soap for 30 seconds, and then rinse hands with tap water for 30 seconds.

For the benzalkonium chloride-based sanitizer and ethyl alcohol-based sanitizer groups, subjects were instructed to refrain from washing hands with soap. Hand cultures were obtained 4 times during each of 41, 3-hour clinic sessions: T₀, 1 minute following antiseptic application; T₁: 1 hour after application; T₂: 2 hours after application; and T₃: 3 hours after application. Samples were collected from subjects at these intervals immediately after leaving a patient’s room, prior to any additional hand hygiene regimens. Samples were collected by taking a 5-second imprint of the middle and index fingers of the dominant, ungloved hand of each subject on commercial contact bacterial agar plates. Plates were incubated at 35°C under aerobic conditions. The total bacterial contamination of hands was recorded as the number of colony forming units (CFU) per plate after 72 hours of incubation. Counts were made to a maximum of 50 CFU due to confluent growth of colonies beyond this point. No anaerobic cultures were performed. The trained microbiologist examining the culture plates was blinded to the hand hygiene method used. All microbiological testing was performed by an independent laboratory, Water-Jel Technologies, LLC (Carlstadt, NJ).

The number of patients contacted and the number of hand hygiene regimens was recorded over the 3-hour time period. Each subject completed a Hand Skin Assessment (HSA) questionnaire, a self-rating scale used to measure skin condition, prior to the trial. Subjects give their dominant hand a score from 1 to 7 in 4 dimensions: appearance, intactness, moisture content, and sensation. The resulting score ranges from 4 to 28, with 28 indicating completely healthy skin. Hand Skin Assessment scores correlate significantly with physiologic measures of skin damage.¹³ The frequency of gloving was also recorded to address potentially confounding effects of glove use on hand microbacterial counts.

Data analysis. Microbial counts were converted to log10 to normalize the data. Statistical analysis was performed using analysis of variance to compare microbial counts between the 3 intervention groups at each time point and to compare microbial counts of each intervention group between time points. Changes in HSA scores were determined for each subject in the trial. Nonparametric analysis was performed using Mann-Whitney U tests and Kruskal-Wallis tests as indicated. All statistical analysis was performed using Stata/MP 13.0 (StataCorp LP, College Station, TX). Threshold for significance was set at P < 0.05 with use of the Bonferroni correction method.

Ninety-five subjects participated in the trial, which took place over 41 separate 3-hour clinic sessions. Thirty-six subjects were randomized to the ethyl alcohol-based sanitizer group (37.9%), 38 subjects to the benzalkonium chloride-based sanitizer group (40%), and 21 subjects to the soap-and-water handwashing group (22.1%). Seventy-five subjects were male (78.9%) and 20 subjects were female (21.0%). The average age of the study population was 28.6 years (median 29 years, range 22-38 years). Sixty-four subjects identified as white (67.4%), 26 as Asian (27.4%), and 5 as Hispanic (5.3%). Prior to the trial, 34 subjects reported having redness, itching, cracking, or dryness of the hands that was mild and infrequent.

So as to ensure exclusion of doubtful high CFU trials, a more restrictive criterion was applied to plates with > 50 CFU or too numerous to count, and those plates were eliminated. Using this criteria, 9 plates in the ethyl alcohol-based sanitizer group, 11 plates in the benzalkonium chloride-based sanitizer, and 10 plates in the soap-and-water handwashing group were excluded from the study. The excluded plates came from variable subjects at each time point. For example, if a plate from a certain subject was excluded because of high CFU counts at T₀ this subject’s plate was not necessarily excluded at subsequent time points. See Table 1 for complete results. 

Significant results from post hoc analysis between time points listed. T₀: 1 minute after application of hand sanitizer; T₁: 1 hour after application; T₂: 2 hours after application; and T₃: 3 hours after application.

Implementing the previous restrictions, the hand bacterial counts for all eligible subjects at each time point were averaged (Table 2). Hand bacterial counts increased significantly over the 3-hour clinic session with ethyl alcohol-based sanitizer (9.24 to 21.90 CFU, P = 0.0021), benzalkonium chloride-based sanitizer (6.69 to 21.59 CFU, P <0.00001), and soap-and-water handwashing (8.43 to 22.75 CFU, P = 0.0009) based on analysis of variance testing. Post hoc subgroup analysis between time points indicated there was a significant difference in CFU between T₀ and T₃ with all hand hygiene methods (Figure 1). With the benzalkonium chloride-based sanitizer, there was also a significant difference between T₀ and T₁ as well as between T₀ and T₂ (Table 2). Interestingly, the distribution of hand bacterial counts appeared bimodal for all hand hygiene methods. The percentage of subject trials with ≤ 20 CFU decreased over the course of the study period for all hand hygiene methods. Conversely, the percentage of subject trials with greater than 20 CFU increased during the study for all hand hygiene methods. The percentage of conversion from low to high colony count trials was lowest for benzalkonium chloride-based sanitizer (29%), compared with ethyl alcohol-based sanitizer (48%) and soap-and-water handwashing (57%). 

The change in HSA score for ethyl alcohol-based sanitizer was 0.25 ± 1.02 (n = 36, median 0, range -3 to 3), benzalkonium chloride-based sanitizer was 0.29 ± 0.93 (n = 38, median 0, range -2 to 2), and handwashing was -0.95 ± 1.43 (n = 21, median 0, range -5 to 0). Nonparametric analysis between the 3 experimental groups indicated a significant difference between the groups (P = 0.0010). Post hoc subgroup analysis revealed both ethyl alcohol-based sanitizer (P = 0.0009) and benzalkonium chloride-based sanitizer (P = 0.0007) resulted in better hand condition than soap-and-water handwashing based on change in HSA score before and after a 3-hour clinic session using a single hand hygiene regimen. Multivariate regression analysis was performed with change in HSA score as the dependent variable and number of hand washings, number of hand dryings, number of patients contacted, and total number of hand hygiene regimens during a 3-hour clinic session as independent variables. None of the independent variables were found to be predictors of change in hand condition (P = 0.1844). 

It is important to note there is a difference between transient flora and resident flora. Transient flora is acquired from patient contact and contaminated surfaces, more frequently associated with health care-associated infections, and more amenable to removal with routine handwashing. Resident flora, on the other hand, is less likely to be associated with infection and is more resistant to removal. The number of both types of flora varies between individuals, but stays relatively constant for each person. There was no transient flora found on the hands of subjects in the authors’ study, reinforcing that this was a monitored study and that when subjects are forced to utilize hand hygiene measures, pathogenic bacteria can be minimized in outpatient clinical settings. In addition, certain subjects who participated in the study more than once were found to consistently have high resident flora counts, which is consistent with the idea that some individuals will naturally have higher bacterial counts than others,¹⁴ and this is not an indication of lack of efficacy of the hand hygiene regimens in this population.

It has previously been shown that duration of clinical activity with patient contact is associated with intensity of hand bacterial contamination.¹⁵ The authors’ study findings are congruent with those findings, as hand bacterial counts increased from the beginning of the trial to the end in all treatment groups. Upon analysis, there was no significant difference between the treatment groups even with the different hand hygiene methods. Despite receiving only 1 application, subjects in the benzalkonium chloride-based sanitizer group had an increased accumulation of bacteria that was not significantly different when compared to the other 2 groups that were utilizing hand hygiene regimens between each patient encounter. Another consideration must be that normal hand flora will proliferate under gloves, which could account in part for the increase in mean CFU count in every treatment group throughout the trial period.¹⁶

Benzalkonium chloride-based sanitizer has persistent activity, which inhibits the proliferation and survival of organisms hours after application. Alkyl benzalkonium chloride is a quaternary ammonium compound that is the active ingredient in the benzalkonium chloride-based sanitizer used in this study. The activity of alkyl benzalkonium chloride is thought to be by cytoplasmic adherence with subsequent lysis of microorganisms. The compound is bacteriostatic and fungistatic, but also bactericidal at high concentrations. There is a relative dearth of data regarding quaternary ammonium compounds in hand hygiene because of its infrequent use in the United States. It has, however, been shown to outperform alcohol-based sanitizers in 1 study.¹⁷ Clearly, additional studies will be helpful in elucidating the role of quaternary ammonium compounds in infection control, including its usefulness in inpatient clinical settings, especially in the intensive care unit, where more than 25% of nosocomial infections occur,¹⁸ as well as its efficacy against transient flora.

The adherence to hand hygiene measures has been directly related to the decreased prevalence of health care-associated infections.^19,20 The results of this study are critical in that they demonstrate a long-acting hand sanitizer is not only as effective as soap-and-water and alcohol-based sanitizers, but also appears to improve hand condition in comparison to soap-and-water handwashing. Considering the lack of compliance with hand hygiene in clinical settings is secondary to inconvenience, lack of time, and skin irritation, a hand sanitizer that reduces the number of required uses and improves hand condition may increase overall compliance, and thus decrease health care-associated infections. Importantly, implementation of a long-acting sanitizer would also potentially reduce costs compared to other methods requiring expensive technology or large numbers of additional personnel to monitor hand hygiene compliance. This is significant considering the cost of health care-associated infections results in an additional $5 billion-$16.6 billion in the United States annually.²¹

The authors would like to thank Verce Tolomanowski, microbiologist at Water-Jel Technologies, LLC, Carlstadt, NJ, for assisting with microbiological testing and cultures related to this study.

Affiliation: Division of Plastic and Reconstructive Surgery, Department of Surgery, Rutgers New Jersey Medical School, Newark, NJ

Correspondence:
Mark S. Granick, MD
Division of Plastic and Reconstructive Surgery
Department of Surgery
Rutgers New Jersey Medical School
140 Bergen Street, Ste. E1620
Newark, NJ 07103
mgranickmd@njms.rutgers.edu

Disclosure: This study was funded by Bioderm Technologies, Inc, Trenton, NJ. The sponsors had no involvement in the study beyond funding.