Tap water versus sterile saline solution in the colonisation of skin wounds

Abstract

Irrigating wounds with tap water does not increase colonisation, but controlled studies are required for further evidence. Microbial colonisation was assessed in skin wounds, before and after irrigation with tap water, and was compared with irrigation using 0·9% sodium chloride sterile solution. The study included 120 subjects with chronic, traumatic, vascular, pressure or neuropathic wounds. A total of 60 wounds were randomly assigned to be irrigated with tap water (tap water group) and another 60 to be irrigated with 0·9% sodium chloride sterile solution (saline group), at a pressure of 0·46–0·54 PSI. Samples were collected from the centre of each wound using Levine's technique, before and after irrigation, and cultivated in thioglycollate, hypertonic mannitol agar, eosin methylene blue (EMB) agar, blood agar and Sabouraud agar at 37°C for 72 hours. There was concordance (kappa test) and discordance (McNemar test) regarding the count of positive and/or negative samples before and after irrigation in each group. The proportion of reduction of positive samples was similar for both groups in all cultures. Colony‐forming unit count before and after irrigation was similar in both groups and in all cultures, except for the culture in hypertonic mannitol agar from the tap water group, for which the count was lower after irrigation (Wilcoxon z = 2·05, P = 0·041). It is concluded that skin wound irrigation with tap water leads to further reduction of Gram‐positive bacteria compared with 0·9% sodium chloride sterile solution, with no difference in colonisation of haemolytic bacteria, Gram‐negative bacteria and fungi.

Introduction

A wound is a solution of continuity of the skin tissue and may be associated with the loss of subcutaneous tissue, muscles and bones 1. The wound may be contaminated when it presents microorganisms without multiplication activity; colonised, when the microorganisms adhere to the tissues with active multiplication, without causing a host response; and infected, when there is invasion of microorganisms in the adjacent healthy tissue, causing an immune response in the host 2. Chronic wounds may be colonised by bacteria 3.

In the wound healing process, it is essential to prepare the wound bed for healing; the first action in this care is cleaning, which can be performed by fluid irrigation, for removing the bacteria and loosely adhered foreign material 4, 5, 6. For this irrigation, 0·9% sodium chloride sterile solution has been the treatment of choice, but tap water is a suitable alternative 6, 7, 8.

Experimental and clinical studies to evaluate the colonisation of skin wounds samples collected before and after irrigation with tap water have presented strong evidence of its effectiveness, as no differences in colonisation were observed when compared with irrigation using 0·9% sodium chloride sterile solution. However, these studies were limited by the difference in volume and pressure of the cleaning fluids, as well as by their small sample sizes 9, 10, 11, 12.

Therefore, because of the need for further investigation on this issue, this study aimed to assess microbial colonisation in skin wounds before and after irrigation with tap water, and to compare it with samples irrigated with 0·9% sodium chloride sterile solution.

Materials and methods

This was a cross‐sectional, controlled, randomised study approved by the local Ethics Committee under Protocol No. 1723/11. A total of 120 subjects agreed to participate and signed an informed consent. Subjects were aged from 18 to 90 years, and most patients were in the age range from 55 to 70 years. The types of wounds observed were chronic wounds of vascular aetiology (62%), traumatic wounds (8%), pressure ulcers (14%) and neuropathic wounds (16%). Wound extension ranged from 2 to 5 cm in length; the wounds were located on the lower limbs (88%) and trunk (12%), and the evolution ranged from 15 to 400 days. Debilitated patients and those who presented wounds of infectious origin or wounds below the subcutaneous tissue, according to medical evaluation criteria, were excluded. The sample size was determined based on a minimum size, which is calculated when a P‐value estimate is not known, according to Levine et al., with a test power of 80% 13.

The subjects were randomly allocated to two groups, using BioEstat version 5.0 (Ayres, Belém, Pará, Brazil), based on a numerical sequence of 1 to 120, assigning 60 numbers to the saline group and 60 numbers to the tap water group. During selection of the subjects, they were sequentially numbered and allocated to the respective groups. In the saline group, a 0·9% sodium chloride sterile solution from Linhamax^® (Segmenta, Ribeirão Preto, São Paulo, Brazil), in 250 ml bottles, was used. Tap water was used in the Tap Water Group, always from the same tap, in the Outpatient Wound Treatment Clinic of a University Hospital, where the present study was conducted. The tap water was assessed using bacteriological analysis, in accordance with the Guidelines for Drinking‐Water Quality of the World Health Organization 14.

To ensure that the team members who performed data collection and analysis were blinded as to the cleaning fluids used, one of the researchers was responsible only for preparing the fluids for irrigation, which were designated Fluid A and Fluid B. The decision on which would be Fluid A or B was made randomly by draw; the result was recorded on a piece of paper, which was placed in an opaque, sealed envelope and kept confidential until the data analysis phase. For the water preparation, this team member disinfected the tap, opened it and let the water run for 10 seconds. Afterwards, prepped with a surgical mask, apron and sterile gloves, this member collected the water in a sterile basin, aspirated its contents with a 50 ml sterile syringe and transferred into a 250 ml bottle, the same one used for the sodium chloride sterile solution, which had been emptied at the same time for the replacement. Afterwards, a bottle with either tap water or sterile solution was attached to a sterile macrodrip parenteral infusion set (Intrafix^®Primeline, B. Braun in Brazil, São Gonçalo, Rio de Janeiro, Brazil), wrapped in a polypropylene plastic bag and identified as Fluid A or B. A microwave oven (Mef30, Electrolux^®, Manaus, Amazonas, Brazil) was used to heat the fluid for 30 seconds at 1000 W to an approximate temperature of 36°C.

For wound irrigation, in accordance with the randomisation sequence, a bottle containing either Fluid A or Fluid B was set on a pedestal tripod, hanging from a height of 210·0 cm from the floor. The edge of the device was kept 15·0 cm from the wound, with the fluid output control device closed. The patient was positioned on an examination table, with the wound over a container. The output control device was opened and the fluid flowed out freely, by gravity, with the stream directed towards the centre of the wound. The driving pressure of the fluid was measured at the exit end of the equipment using a universal meter (Biotek^®, model DPM.2 plus, series No. 120508; Biotek, Winooski, VT): 0·24 mmHg (0·46 PSI) at the beginning of the flow and 0·28 mmHg (0·54 PSI) at the end of the flow, in both groups. Two samples of biological material were collected from the centre of the wound, one before and another after irrigation, using Levine's technique 15. The swab was placed in a test tube with 1 ml of 0·9% sodium chloride sterile solution and sent to the microbiology laboratory. The wound was treated according to each individual prescription.

In the microbiology laboratory, three team members were responsible for the culture and measurements, with agreement between them. With the same swab, suspension seeding was performed in thioglycollate broth in a vial, to assess the growth of microorganisms. Then, to reduce colony‐forming units (CFUs) and allow for its measurement, each sample was diluted to 10⁻³. Afterwards, 0·1 ml of the sample was placed in four Petri dishes, with the respective culture media: Sabouraud agar with chloramphenicol, selective for fungi; hypertonic mannitol agar, for Gram‐positive staphylococci; eosin methylene blue (EMB) agar, for Gram‐negative bacilli; and blood agar, for haemolytic microorganisms. The Petri dishes were then incubated at 37°C for 72 hours. For the assessment, the plates that presented positive microorganism culture were separated and counted, the CFUs were counted and the result was multiplied by 10³, to compensate for the dilution. The differentiated CFUs were harvested with a platinum wire loop for inoculation, stained using the Gram stain method and identified in an Olympus^® optical microscope (Olympus Optical do Brazil, São Paulo, Brazil) with 1000× magnification.

At the end of 5 months, the total sample of 120 wounds was reached, with 240 collections of biological materials and 1200 cultures. For each group, the proportion of agreement between positive and negative samples before and after irrigation was assessed by the Kappa test. Discordance between positive and negative samples was assessed by the McNemar test and the proportion of reduction of positive samples between both groups, by Fisher's exact test. For each group, the CFU count in culture media, before and after irrigation, was compared using the Wilcoxon test. The level of rejection of the null hypothesis was set at 5%, considering significant P‐values less than or equal to 0·05 16. Statistical analysis was conducted using the Statistical Package for the Social Sciences (SPSS) 20.0 for Windows (SPSS Inc., Chicago, IL), by statistical analysts who did had not participated in the previous steps.

Results

Concordances and discordances in positive and/or negative samples, before and after irrigation, were observed equally in both groups and in all culture media, as evaluated by the Kappa and McNemar tests (Table 1).

Table 1.

Culture media in the tap water and saline groups before and after irrigation

Culture medium	Groups	Concordance samples (%)	T. Kappa concordance	Discordance samples (%)	T. McNemar discordance	Reduction of samples positive (%)	T. Fisher water versus saline (P)
Thioglycollate	Tap water	90	z = 4·76	10	P = 0·6875	3·3	0·3392
			P = 0·0001
	Saline	90	z =5·49	10	P = 0·6875	6·7
			P < 0·0001
Blood agar	Tap water	90	z = 5·9761	10	P = 0·6875	3·3	0·2195
			P < 0·0001
	Saline	86·7	z = 5·5392	13·3	P = 0·7266	8·3
			P < 0.0001
Hypertonic mannitol agar	Tap water	82	z = 4·9367	18	P = 0·5488	12	0·3811
			P < 0·0001
	Saline	83	z = 5·0564	17	P = 1·000	0·8
			P < 0·0001
EMB agar	Tap water	85	z = 5·4030	15	P = 0·5078	5	0·5000
			P < 0·0001
	Saline	88·3	z = 5·7852	11·7	P = 1·0000	6·7
			P < 0·0001
Sabouraud agar	Tap water	100	z = 0·90	0	P = 1·000	0	0·2479
			P = 1·000
	Saline	96·6	z = 4·3957	3·4	P = 0·5000	3·4
			P < 0·0001

The proportion of reduction of positive samples after irrigation was similar in both groups in all culture media, as assessed by Fisher's exact test (Table 1).

There was no difference in the CFU count before and after irrigation, for any of the groups or culture media, except for the hypertonic mannitol agar, for which the difference was significant (lower CFU count in the tap water group), as assessed by the Wilcoxon test (Table 2).

Table 2.

Cultivation media in the tap water and saline groups, with measurement of colony‐forming units (CFU) before and after irrigation (N = 60 in both groups)

Culture medium	Groups	Colony‐forming units		Wilcoxon test
		Before cleansing	After cleansing
Blood agar	Tap water	169500·00	159000·00	z = 1·06
				* P = 0·291
	Saline	149333·33	135500·00	z = 1·29
				* P = 0·197
Hypertonic mannitol agar	Tap water	108666·67	89500·00	z = 2·05
				**  P = 0·041
	Saline	78666·67	71000·00	z = 0·63
				* P = 0·531
EMB agar	Tap water	25000·00	5000·00	z = 1·0
				* P = 0·3173
	Saline	0	0	–
Sabouraud agar	Tap water	0·4333	0·1333	z = 1·6036
				* P = 0·1088
	Saline	0·2500	0·0333	z = 0·4472
				* P = 0·6547

^*Value indicative of similarity.

^**Value indicative of difference.

Discussion

The results of the present study demonstrate that colonisation of haemolytic, Gram‐positive and Gram‐negative bacteria and fungi was similar in both groups, as evidenced by the concordance on the presence or absence of microorganisms in the wound samples and on the percentage of their reduction, when comparing both fluids before and after irrigation. The CFU count also evidenced that Gram‐negative bacteria, haemolytic bacteria and fungi counts were similar before and after irrigation in both groups, while the Gram‐positive bacteria count differed, indicating that irrigation with tap water had significantly reduced it.

Experimental studies with a similar methodology presented similar results; some observed even better results with the use of tap water 9, 10, 11. Moscati et al. analysed samples collected before and after irrigation with both fluids for 4 minutes, in 10 wounds on skin of rats, per group, inoculated with Staphylococcus aureus. A statistically significant difference was observed, with further reduction of bacteria, in wounds irrigated with tap water, but fluid volume and pressure were not controlled 9. Svoboda et al. analysed musculoskeletal wounds of seven goats per group, which had been inoculated with luminescent bacteria, and assessed the photon count of the images obtained from samples collected before and after pulsatile irrigation with 9 l of both fluids. They observed a significant difference in bacteria when using both fluids, without any quantitative variation between them 10. Resende et al., also in an experimental study, albeit longitudinal, irrigated 20 wounds (2·4 cm diameter) on the backs of 40 rats with 150 ml of each fluid daily, for 7 days, with controlled pressure. Samples were taken every 2 days and cultured. They demonstrated that there was no difference between colonised and non‐colonised cultures, nor in Gram‐positive and Gram‐negative bacteria, haemolytic bacteria and fungi counts between both groups 11. The present study is a follow‐up to that study; it was conducted at the same institution, and water from the same tap was used.

A clinical study, also similar to the present one, but with a sample of 46 individuals (12), also observed no difference in the CFU count before and after irrigation with both fluids. The present study, with 120 adults, is comparable to that one, presenting an advantage in favour of tap water because of the smaller growth of Gram‐positive bacteria 12.

The similarity in the colonisation of wounds irrigated by both fluids indicates that tap water, although not sterile, does not lead to contamination and subsequent colonisation. The only difference observed – the reduction of CFU count of Gram‐positive bacteria after irrigation with tap water – may be related to the sodium chloride rather than to the water. The hypertonic mannitol agar culture medium promotes the growth of Gram‐positive bacteria and is rich in sodium. Therefore, it can be concluded that irrigation with 0·9% sodium chloride sterile solution provided a favourable environment for the survival of these bacteria, unlike tap water, which had the advantage that it led to a greater reduction of this species.

The results indicate that fungal growth was insignificant, similar in both groups, as it is not common to find fungi on wounds such as those analysed. A small growth of Gram‐negative bacteria was observed in the tap water group, and no growth was observed in the saline group. In all these situations, there was a reduction of microorganisms after irrigation. The observed Gram‐negative bacteria could have come from the digestive tract, through faecal evacuation, contaminating the wounds during body hygiene; irrigation contributed to its reduction.

The results of this study consolidate the previously verified premise that wound irrigation removes bacteria and other microorganisms that could lead to infection 4, 6. Wound cleansing is the act of removing foreign material loosely adhered to its bed; it can be performed by means of irrigating fluids, and it should be performed at each change of dressing 4.

A continuous sequence of contamination, colonisation and infection of the wound is now widely accepted, and colonisation is the phase that precedes the infection 5. Infection is feared by health care professionals and patients with wounds. However, literature reviews have retrieved studies that also observed no differences in infection of wounds irrigated with both fluids; these reviews indicated that some of these studies had biases that could affect the results, but pointed to the possibility that the use of tap water is more advantageous 7, 8. Authors of a more recent clinical study that assessed the infection rates of wounds irrigated with both fluids reported its random design and control of irrigation technique and irrigated volume as strengths, as well as the fact that it was a blinded study with a relatively large number of individuals − 625. However, it still had limitations, such as the fact that fluid temperatures were not identical and that subjective indicators of infection were used; the authors justified the absence of cultures and bacterial count because of practical difficulties associated with the nature of the study 17. Therefore, although the present study did not assess wound infection in a prospective follow‐up, it adds to previous studies and can contribute to supporting the use of tap water for cleaning wounds.

The fear of using tap water to clean wounds is because of uncertainty as to whether this practice will contaminate the wound, increase colonisation or even infect it. Unlike the use of sterile solutions, which is widely accepted, this is not yet an accepted practice for many professionals and health care institutions.

The alternative use of tap water for wound cleaning offers professionals and health care institutions an opportunity to reduce the cost of dressings and facilitate access, as it is universally available. However, it must be drinking water, with quality control.

Limitations of this study include the collection of material in a single phase, without monitoring the evolution of colonisation or assessing the clinical signs of infection.

This study aims to expand prospective research on the use of tap water, assessing colonisation and infection in wounds of the same type, but with greater depth, including muscle tissue, bones and cavities.