A critical review of irrigation techniques in acute wounds

Abstract

Irrigation is a fundamental concept to the care of acute wounds and therefore of prime importance in all surgical specialties. A significant amount of research has been done in search of the most effective irrigation technique over the last three decades. There is evidence to show that irrigation is beneficial in the management of acute traumatic and elective wounds, but exactly which techniques and irrigation pressures needed for optimal outcome are still undetermined. A significant number of the studies to date lack the rigid scientific design needed in a climate of evidence‐based medicine. There are substantial methodological flaws and a lack of standardisation between many studies hindering the ability to draw scientific conclusion and compare different studies. There are few randomised, controlled studies on human wounds, and extrapolation from animal studies must be done with caution. Investigation into different irrigation pressures for different levels of contamination is lacking. This would possibly produce a more logical approach for clinically managing acute wounds rather than a blanket treatment for all acute wounds irrespective of the level of contamination. Further research in this direction with greater attention to scientific methodology and design is needed.

Introduction

Wound irrigation is a cornerstone of wound management (1), which endeavours to remove devitalised tissue debris, dirt and bacteria (2), bearing in mind that all traumatic wounds can be categorised as dirty wounds (3). Irrigation of particles such as fat detritus increases the rate of bacterial clearance, which would otherwise have to be phagocytosed and removed by leucocytes (4). Healing accelerates by prior removal of bacteria, haematoma and other contaminants from the wound (5). The concept of irrigation in the management of acute wounds (traumatic and elective) is divided into irrigation technique (pressure related) and the irrigant used. This review discusses irrigation techniques and the potential benefits and harm that they have on patient outcome in regard to soft tissue elective and traumatic wounds.

Discussion

Elective and traumatic wounds are acute wounds that heal by haemostasis, inflammation, new‐tissue formation and remodelling (6). Irrigation has been defined as ‘the process of washing a wound’ with a ‘solution’ (7). In a review by Lawrence (8), a ‘gentle mop or wipe’ is described as a method of irrigation. It is important to make the clear distinction that this technique is another aspect of wound cleansing and is not strictly irrigation. Singleton et al. (9) showed that irrigation with normal saline solution reduced wound infection rates and that effectiveness was proportional to the volume of solution used. This showed that irrigation conferred benefit to wound healing and therefore patient outcome. However, the ideal technique is still undecided, despite over 30 years of research.

Gross et al. (10) investigated the effect of pulsating water jet lavage on experimental contaminated rat wounds. A vertical incision in the masseter was made, 1·5 cm long, and extending to the lateral surface of the mandibular ramus. Each wound was contaminated with 0·03 ml of moist soil containing equal concentrations of Staphylococcus aureus, Proteus mirabilis, Pseudomonas aeruginosa and Klebsiella pneumoniae. The total number of bacteria in each sample was approximately 3 × 10⁷. However, ‘nearly equal soil samples were introduced into the wounds’ suggests some disparity in bacterial count between wounds. The conclusion that pulsating jet irrigation showed a statistical significant reduction in the incidence of infection compared with conventional bulb syringe irrigation must be challenged. They report wounds swabbed with positive culture results, as being infected. Although the degree of contamination occurring during surgery may indicate whether wound infection is likely to occur (11), Gross et al. (10) have repeatedly stated that the wounds were infected without any reference to the clinical state of the wound or their definition of infection such as that defined by the National Research Council (12). It is likely that the word ‘infected’ should more accurately have been ‘contaminated’. There‐ fore, the results suggest that pulsating jet lavage may be beneficial in reducing the likelihood of postoperative infection. Although well controlled, this experiment used rats not humans, and hence, results cannot be used with direct reference to human wounds such as ‘combat wounds’ that the authors state. Sampling error due to the jet of water pushing the bacteria deeper than the conventional bulb syringe was not addressed.

Carlson et al. (13) assessed different pulsed lavage irrigating tips (single orifice, multijet and radiant) and varying irrigating pressures. This controlled study on open dog leg wounds used radiographic and histological analysis to assess dispersion of radiopaque material and to detect fibre separation, respectively. The wounds were 5 mm deep except for those irrigated with the radiant catheter tip (20 mm deep), and no explanation for this discrepancy is given. Using the multijet tip, there was no spread of liquid beyond the debris originally present in the wound nor was there any fibre separation detected at pressures from 30 to 90 psi. The single jet tip and radiant catheter tip produced dispersion and fibre separation particularly at higher pressures. Although this suggests that the multijet tip is less traumatic to the tissues, the experiment fails to quantify the total amount of debris left in each wound after irrigation, which is critical in contaminated wounds. The reduction in debris could be substantially more using the single jet tip and radiant catheter tip, which could then confer overall benefit. Water inclusion did not occur with any of the irrigation delivery tip and pressure combinations tested. This study suggests that acute human wounds may benefit from reduction in tissue trauma with less dispersion of debris and fibre separation when the multijet tip is used, but the quantification of debris reduction has yet to be assessed.

Although the effect of pulsating jet lavage was previously compared with bulb syringe (10), this was using simple laceration wounds and not crush‐type wounds that are a common type of traumatic injury. Gross et al. (5) compared the effectiveness of pulsating water jet lavage (70 psi) with bulb syringe in crushed, contaminated wounds using rats. A vertical incision in the masseter was made, 1·5 cm long, and extending to the lateral surface of the mandibular ramus. A second parallel incised wound was made in the muscle. Each wound was contaminated with 0·03 ml of moist soil containing equal concentrations of S. aureus, P. mirabilis, P. aeruginosa and K. pneumoniae. The total number of bacteria in each sample was approximately 3 × 10⁷. Pliers were used to crush the parallel incisions with a contact area of 4 × 6 mm. This well‐controlled experiment actually looked at inflammatory reaction, abscess formation and amount of foreign material present in the wound as well as number of positive cultures and bacterial density scores, which their previous study had failed to do. The results indicated that pulsating jet lavage was more effective in reducing bacterial population, removal of necrotic tissue and foreign particles. A correlation was demonstrated between the amount of inflammatory reaction and abscess formation and the amount of foreign material in the wounds. The authors state that early reduction of bacterial concentration by pulsating jet lavage resulted in accelerated wound healing. The validity of this statement is unsupported by the data, as the time taken for the wounds to heal was not measured.

Rodeheaver et al. (3) investigated guinea pig wound cleansing by continuous high‐pressure irrigation. Their study involved the concept of infection potentiating factors, components of sterile soil which exert damaging effects on tissues. Montmorillonite, an infection potentiating factor, was chosen to contaminate the wounds. Incised wounds parallel to the vertebral column were made in the guinea pigs. No depth was described in this article. One hour after 20 mg of soil containing montmorrillonite was deposited in each wound, irrigation at 1, 5, 10 and 15 psi was carried out. Ten minutes after irrigation 10³ S. aureus was added to each wound and microbiological assessment carried out 4 days after. The authors do not state that the soil used is sterile and rely on the indirect methods (potentiating infection) to detect its presence in the wound after irrigation. This compromises the validity of the results, as the infections could be a result of bacteria originally from the soil and not the infection potentiating factors. High‐pressure irrigation of 10–15 psi significantly increased the amount of soil removed (75–84%). The authors state that fluid dynamics of a high‐pressure irrigation stream (continuous or pulsatile) do not influence the efficacy of decontamination. This is based on previous experience with pulsatile and continuous irrigation at identical pressures which were apparently equally effective at bacteria removal; however, the authors fail to provide any data to support this statement.

The possibility of tissue injury, dissemination of surface foreign bodies throughout the wound and other theoretical side‐effects of using high‐pressure irrigation were investigated by Wheeler et al. (2). A 35‐ml syringe with 19‐gauge needle and pulsatile irrigator delivering 8 and 70 psi, respectively, were used at 3 cm from pig wounds. The incised wounds were made on the back, 3 cm long and 0·5 cm deep. Each wound was inoculated with 10³ Serratia marcescens, and no soil was used. The results showed significantly more lateral fluid dissemination with the pulsatile device. The distribution of bacteria using pulsatile or syringe, however, was the same as the control group but with a markedly reduced count. Interestingly, the magnitude of decontamination by syringe or pulsatile irrigation did not differ significantly although the pressures were significantly different (8 and 70 psi). Control pressure was not quantified reducing the validity of the results. High‐pressure syringe and pulsatile irrigation was shown to impair ability of wounds to resist infection compared with the control. ‘Fears that high‐pressure irrigation of the wound will disseminate contaminants into adjacent tissue appear to be unfounded’ is an invalid conclusion which must be challenged. This experiment only assessed bacteria and not other types of contaminant, therefore limiting the use of the study. Larger contaminant particles other than bacteria may disseminate at high pressure. This limitation has not been recognised by many subsequent authors. However, it is apparent that high‐ pressure syringe or pulsatile irrigation is likely to be beneficial to reduce bacterial count, dirt and tissue debris in heavily contaminated wounds outweighing the risk of tissue injury. In relatively clean wounds, the damage of tissue defences from high‐pressure irrigation would appear to outweigh the benefits, being potentially harmful.

A survey of the management of acute traumatic wounds by emergency physicians in the USA (14) showed that 75% of providers irrigated wounds, however, many of the irrigation techniques used, had not been proven to generate the minimum of ‘8 psi considered appropriate for tissue cleaning (Wheeler et al.)’. Bulb syringe was used by 15% and syringe with plastic catheter by 85%.

Chisholm et al. (15) carried out a prospective, randomised, controlled study on patients presenting with traumatic wounds to the emergency department (ED), comparing the use of 220‐ml saline pressurised canister (8 psi) with 250 ml of saline syringe irrigation per 5 cm length of laceration. The lack of standardisation in volume of irrigant and antibiotic ointment used reduces the validity of the results. Irrigation times were significantly reduced (mean 3·9 minutes versus 7·3 minutes, respectively). Syringe method was labelled labour intensive. Wound complication was 5% and 3·6% for canister and syringe irrigation, respectively, but not significant (P = 0·5). The authors endorsed the pressurised canister, as it generated adequate irrigation pressures and was not operator dependent. Animal studies (2) have shown that high pressures (8 psi) can cause damage to tissue defences although this cannot be automatically extrapolated to humans. As irrigation proceeds, the level of contamination may decrease requiring lower‐pressure irrigation; hence, constant pressure may be disadvantageous. Additionally, wound contamination is not uniform, and areas that are less contaminated from the outset could be irrigated with lower pressure. In this respect, the operator has less control with the canister. For research purposes, the pressurised canister would produce standardisation.

Singer et al. (1) compared irrigation pressures attained by human volunteers, using various techniques. Unfortunately, impact irrigation pressure on the wound surface was not measured, only that at the pressure transducer in a closed system, not representative of clinical wound irrigation. A small sample size (ten volunteers) was used, reducing the power of the results. All ten volunteers were men, not representative of professionals irrigating wounds. A further flaw in this study was lack of standardisation in regard to using one or two hands to perform the irrigation. Additionally, apparatus used to measure was modified, but no reference was made to validation of the modification. The results showed that the median peak irrigation pressure generated with a 35‐ml syringe on a 19‐gauge needle was 35 psi. Using a 65‐ml syringe on a 19‐gauge needle, this pressure was 27·5 psi. Using either IV bag or plastic bottle pierced with 19‐gauge needle failed to generate the irrigation pressures (10–15 psi) recommended by Rodeheaver et al. (3). Using IV bag inside a pressure cuff at 400 mmHg infused via a 19‐gauge needle produced a median peak pressure of 10 psi, but this doubled the irrigation time. Using the pressure cuff with a 16‐gauge needle reduced not only the duration but also the pressure to inadequate levels. The results suggest that 35‐ or 65‐ml syringe with 19‐gauge needle would be rapid and effective for irrigation at high pressures, benefiting highly contaminated wounds. Additionally, as the authors note, this apparatus allows closer contact with the wound so that impact pressure does not decrease as much as with bags or bottles that are further from the wound. High‐pressure irrigation can produce splatter and spread of bacteria to the surroundings including other patients and surgeon. The use of a plastic shield at the end of the irrigating syringe reduces this hazard (16). Using universal precautions such as a gown, mask and visor is also advisable.

Morse et al. (17) evaluated the wound infection rate and irrigation pressure of two potential new wound irrigation devices, the Port and Cap. The clinical portion of the study was prospective and randomised. The results showed the infection rates after irrigation to be 1 and 2% for the port and cap, respectively. The authors compare this with a rate of 5–10% per year for all ED wounds in the USA. However, this comparison is not valid as the authors' study excluded ED patients referred to another specialty such as Orthopaedics or Surgery. The low infection rate may be due to this selection bias. More than 50% (110 of 208 patients) had a telephone conversation follow up to detect infection, compromising the reliability of the results. Patients' limited understanding and experience of wound care may have allowed for infections to go unreported. The second part of the study, the experimental phase, compared the pressures and speed of irrigation, comparing cap, port, syringe/needle and syringe/catheter. These results showed pressures no greater than 2·01 psi with the port or cap. The syringe/catheter and syringe/needle had pressures of 8·16 and 7·33 psi, respectively. Unlike other previous studies, this study calculated the pressure of irrigation fluid at the wound surface. This was done using Bernoulli's equation. This equation involves ‘rho’ which equals the density of water. However, the irrigating fluid was not water but saline. This flaw in research design reduces the construct validity of the pressure results, although the significance of this would need to be calculated. The only significant benefit of the new methods was the significantly reduced irrigation time compared with the syringe methods. However, pressures of no greater than 2·01 psi would be insufficient to significantly reduce heavily contaminated wounds (2, 3), rendering these two new devices of little practical clinical use and potentially of more harm then benefit.

Pronchik et al. (18) compared syringe/catheter with the newer ‘port’ method in which a port device is spiked into a bag of saline, on rat wounds. The authors describe these methods as ‘the traditional higher pressure, lower‐volume and “lower pressure, higher‐volume” ’, respectively. This appears to be a description not mentioned in any of the articles discussed so far. Incised paravertebral wounds were made 2·5 cm long, extending to the muscle fascia. The wounds were inoculated with 50 ml of (10⁶ colony forming units/ml) S. aureus. A mean irrigation pressure of 8·8 and 1·6 psi for the syringe and port method, respectively, was demonstrated, in keeping with findings by Morse et al. (17). However, the bacteria washed out was similar for both techniques generating a psi of 8·8 and 1·6. The authors suggest that the effects of both methods are equally effective at washing out bacteria from wounds. The lack of standardisation of the volume of fluid used with the port method (more than double that used with the syringe) produces two variables, device and volume, compromising the construct validity. Had the volume used in the syringe been doubled to that used with the port device, the bacteria washed out may have been significantly more (7). The port method was faster than the syringe catheter method and the author endorses the port technique. The conclusion drawn is not valid, and standardisation with one variable would be necessary to come to a conclusion regarding superiority over the ‘traditional syringe and catheter technique’. The wounds were not contaminated with particulate matter such as soil, and therefore, no conclusion can be drawn regarding the efficacy of the port in removing contaminant particles. Research design was further flawed, using water density in the Bernoulli's equation to calculate saline irrigation pressures.

Cervantes‐Sanchez et al. (19) investigated the use of syringe pressure irrigation to subdermic tissue after appendectomy, as a means of reducing postoperative wound infection. The study was randomised, and the clinician following up was blind to the treatment given. The number of patients required for statistical significance was calculated prior to recruitment. The numbers recruited and included exceeded the 133 patients needed per group adding to the power of the study. Both control and experimental group received prophylactic antibiotics. A 20‐ml syringe with 19‐gauge catheter and 300 ml of saline was used in the experimental group. Cases were further divided into uncomplicated and complicated (gangrene, abscess and diffuse peritonitis). The results showed a significant reduction (P < 0·000001) in postoperative wound infection in complicated cases receiving syringe irrigation. Although uncomplicated cases also had a reduced rate of infection, it was not statistically significant (P = 0·095). The authors misrepresent the results of Wheeler et al. (2), claiming 8–25 psi is not high enough to increase tissue damage. This clinical study shows that syringe irrigation confers benefit to wound healing in complicated appendectomy cases, reducing infection rates.

Conclusion

Selection of an irrigation technique most beneficial and least harmful to patient outcome based on the current evidence available above is not possible. As has been discussed, there are substantial methodological flaws and a lack of standardisation between studies. No author defined high and low pressure. There were only two randomised, ‘controlled’ studies involving human wounds, and extrapolation from animal studies must be done with caution (see Appendix). Many of the studies on human wounds used results of animal studies as a gold standard. The evidence does suggest that irrigation as a concept confers benefit to the patient. It would seem logical to investigate different pressures for each level of contamination. The level of contamination could be graded according to the ‘Wound Classification System’(20). The benefit of high pressure may outweigh the harm of tissue damage in grade 4 wounds (dirty), and the benefit of low pressure in grade 1 wounds (clean), may outweigh the harm of decreased ability to reduce contamination.

Acknowledgement

This review was undertaken as an assignment for my MSc degree course in Wound Healing and Tissue Repair, Wound Healing Research Unit, Wales College of Medicine, Biology, Life and Health Sciences, Cardiff University, UK.


 Summary of irrigation technique studies, in acute wounds. 


Reviewed paper	Study design	Follow up	Major findings
Gross et al. (10)	Irrigation of bacterial‐contaminated incised rat wounds with pulsating water jet lavage versus conventional bulb syringe irrigation. Microbiological assessment	Rats euthanised on 2nd, 4th, 6th, 8th and 10th days, wounds opened and swabbed for culture	Significantly more negative cultures after pulsating jet lavage than after bulb syringe irrigation. Wounds clinically not assessed
Carlson et al. (13)	Irrigation of iron file‐contaminated open dog wounds with radiopaque liquid using pulsed lavage device with three different tips at various pressures. Radiological and histological assessment	Immediately radiological and histological assessment of wounds looking for dispersion of radiopaque material and muscle fibre separation	Pulsed lavage using the multijet tip at pressures 30–90 psi does not produce dispersion of material or fibre separation. No evidence of intercellular water inclusion from any of the methods used. Quantification of debris reduction not assessed
Gross et al. (5)	Irrigation of bacterial‐contaminated incised and crushed rat wounds, with Pulsating water jet lavage verses conventional bulb syringe irrigation. Microbiological and histological assessment	Rats euthanised on 4th, 6th, 8th, 10th, 12th days, wounds opened and swabbed for culture, wound then excised and fixed in formalin for histological analysis	Significantly more negative cultures and lower bacterial concentrations after pulsating jet lavage than after bulb syringe irrigation. Jet lavage appeared more effective in removing foreign particles from wounds than bulb syringe, but no statistical significance to support this finding is given
Rodeheaver et al. (3)	Irrigation of montmorillonite‐contaminated incised guinea pig wounds with continuous irrigation at pressures of 1, 5, 10 and 15 psi. Post‐irrigation contamination with bacteria Microbiological and gravimetric assessment	Gravimetric assessment of soil removal after irrigation. Microbiological assessment of wounds 4 days after bacterial contamination	Continuous high pressures of 10–15 psi removed approximately 80% of soil from wounds compared with approximately 50% when irrigated by 1–5 psi
Wheeler et al. (2)	Irrigation of bacterial‐contaminated incised pig wounds, with pulsatile (8 psi) and syringe irrigation (70 psi). Microbiological assessment	Dissemination of fluid and bacterial quantification in the wounds assessed 4 days after irrigation. Resistance to infection after irrigation also assessed	Pulsatile irrigation at 70 psi causes significant lateral fluid dissemination. Bacterial dissemination did not occur in either method of irrigation. Magnitude of bacterial decontamination was not significantly different for either method. Both pulsatile and syringe caused tissue trauma rendering the tissue more susceptible to infection
Chisholm et al. (15)	Prospective, randomised, controlled study on human wounds in patients presenting to the emergency department. Pressurized canister of (8 psi) versus syringe irrigation	Irrigation times were recorded. Follow up by outpatient clinic, questionnaire or telephone, to assess for wound complications	Irrigating time was significantly reduced with the pressurized canister compared with the syringe. However the volume used for the canister was consistently smaller. No significant difference in wound complications was demonstrated
Singer et al. (1)	Irrigation pressures attained by human volunteers, using various techniques was assessed using a pressure transducer in a closed system	Not applicable – in vitro study	35‐ or 65‐ml syringe with 19‐gauge needle would be rapid and effective for irrigation at high pressures (35 and 27·5 psi, respectively), benefiting highly contaminated wounds. However small sample size, volunteers not representative of health care providers and no standardization regarding use of one or both hands to irrigate
Morse et al. (17)	Prospective, randomised study: (1) Clinical phase – comparing Port/bag and Cap/bottle irrigation on human wounds in patients presenting to the emergency department. (2) Experimental phase – comparing Port, Cap, syringe/needle and syringe/catheter irrigation	Follow up after 2 days by outpatient clinic or telephone to assess for wound infection. Speed and irrigation pressure were measured in the experimental phase	1 and 2% infection rate Port/bag and Cap/bottle irrigation, respectively, but selection bias. Irrigation speed was faster with Port and Cap than the syringe/needle or syringe/catheter. However the Port and Cap both produced not greater than 2·01 psi, which would be insufficient for heavily contaminated wounds.
Pronchik et al. (18)	Irrigation of bacterial‐contaminated incised rat wounds with syringe/catheter versus Port/bag method. Animal and experimental phase	Microbiological assessment and measurement of irrigation time and pressure	Syringe/catheter and Port/bag method are equally effective at washing bacteria from wounds. Irrigation pressure of 8·8 and 1·6 psi for the syringe and port method respectively was demonstrated. However, the Port/bag method used a greater volume of fluid compromising the construct validity of the study. Removal of particulate matter was not assessed
Cervantes‐Sanchez et al. (19)	Randomised, controlled study to evaluate irrigation of human appendectomy wounds on the incidence of postoperative infection. Controls received prophylactic antibiotics. Experimental group also received syringe/catheter irrigation	Assessed clinically in outpatient clinic 2–4 weeks postoperative by evaluator blind to random assignment	Significant reduction (P < 0·000001) in postoperative wound infection in complicated appendectomy cases receiving syringe irrigation compared with no irrigation