Microbiological Identification and Resistance Profile of Microorganisms in Pressure Injuries After the Use of Polyhexamethylene Biguanide: A Series of Fourteen Cases

In April 2016, the National Pressure Ulcer Advisory Panel (now the National Pressure Injury Advisory Panel [NPIAP]) released an updated staging system in which the term pressure injury (PI) replaced the term pressure ulcer.¹ This change was made to standardize the nomenclature of wounds. The wound classification system was updated as well. Members of the Brazilian Association of Enterostomal Therapy (SOBEST) and the Brazilian Association of Dermatology Nursing (SOBENDE) confirmed and translated the document into Portuguese.¹

The NPIAP classifies PI as “localized damage to the skin and/or underlying soft tissue, usually over a bony prominence or related to a medical or other device.”¹ Intrinsic and extrinsic factors that contribute to the occurrence of PI include age, nutritional status, tissue perfusion, hydration, mobility, level of consciousness, comorbidities, friction, shear, humidity, and pressure.²

In Brazil, the incidence and prevalence of PIs are concentrated in hospitalized patients of advanced age (≥60 years) and in patients with restricted mobility and of advanced age who live in long-term care facilities.^3,4 The incidence of PI in patients 60 years and older is 6.5%, compared with 11.0% in the total population.⁵ The prevalence of PI among older adults is as high as 25.6%.³ The prevalence of PI is lower in the United States (4%–14%) than in Europe (18%–20%), Brazil (25.6%), and Canada (36.8%–53.2%).⁵

Studies show that PIs are commonly infected by gram-positive and gram-negative bacteria such as Staphylococcus aureus and Pseudomonas aeruginosa, as well as by bacteria in the Enterobacteriaceae family, including Proteus mirabilis, Citrobacter spp, and Klebsiella spp.^6,7 Because these organisms have demonstrated resistance to antimicrobial agents, antiseptic agents such as polyhexamethylene biguanide (PHMB) have been used for treatment. These antiseptic agents have a chemical structure similar to substances produced by keratinocytes and neutrophils, which contribute to reepithelization and have regenerative and antiseptic properties.^8,9

This research is important to determine whether PHMB has antimicrobial effectiveness and is useful in wound care. The objective of the current study was to evaluate the effectiveness of PHMB on microorganisms in PIs in patients admitted to a university hospital in Rio de Janeiro, Brazil.

This was a single-center, prospective case series evaluating 14 patients with PI receiving initial treatment with PHMB between October 2019 and June 2020 in a large university hospital in Rio de Janeiro. The sample was sequenced, and all people of an accessible population (ie, those who had a PI and were eligible to use the product) who met the eligibility criteria in the specified time period were recruited.¹⁰

During the 10-month study period, there were only 14 PIs in which PHMB had not been previously applied. Thus, the authors of the current study worked with the total population of injuries, justifying the small number in terms of external validity. The sample size was significant in terms of internal validity, however, reaching a power of 0.8.

The inclusion criteria were age 18 years and older, male and female sex, presence of PI, and PIs with no prior application of PHMB. Pregnant women, people allergic to any component of PHMB, patients with a stage 1 PI (per NPIAP¹), patients with lesions in the mucous membrane, and patients with wounds that lacked granulation tissue were excluded.

The study outcome variable was the identification of microorganisms in the lesions. The explanatory variable was the use of PHMB.

Wounds were characterized clinically using a research instrument that measures the amount of epithelialization tissue, granulation, slough, coagulation, and necrosis, based on the percentage of their presence in the wound bed and perilesional area. The percentages are categorized as follows: 1% to 25%, 25% to 50%, 50% to 75%, and 75% to 100%. Quantity of wound exudate was evaluated by observing the dressing; exudate on 25% of the gauze area was considered small, on 50% to 75% was considered moderate, and on greater than 75% was considered large.

Pain was assessed using the 10-point visual analog scale (VAS), which ranges from 0 (no pain) to 10 (severe pain). Because of the possibility of analgesic medication use among the patients during data collection, pain intensity was assessed within 24 hours before data collection began. Odor was evaluated by the researcher’s sense of smell and was classified as present or absent.

After the participants were screened, a short individual questionnaire was administered by a trained nurse to collect information on the patients’ clinical condition and injury characteristics, after which a swab sample was collected from the wound bed, using the Levine technique.¹¹ The wound was cleaned with 0.9% saline, and PHMB was applied according to the manufacturer’s recommendations. A second wound swab was collected 15 minutes after application of PHMB. The microorganisms identified before and after application of PHMB as well as the resistance profile of the microorganisms were evaluated.

The wound swabs were placed in Stuart transport medium and then seeded in 2.0 mL of Brain Heart Infusion Medium and incubated at 35°C (±2°C) for 24 to 48 hours. After the incubation period, the tubes that presented cloudy culture medium were seeded in salted mannitol agar and 5% sheep blood agar and incubated at 35°C (±2°C) for 24 to 48 hours. All plates were analyzed for growth and colony characteristics.

In the next step in the process, samples were identified with the Vitek 2 system (bioMérieux). Susceptibility to antimicrobial agents was analyzed using the disk diffusion method in solid medium, and the results were analyzed per Clinical and Laboratory Standards Institute (CLSI) guidelines.¹²

The tryptone soya agar strains were sown and then incubated at 37°C for approximately 24 hours. Isolated colonies were then transferred to tubes with 2 mL of 0.9% saline until they reached turbidity equivalent to 0.5 on the McFarland scale. The confluent suspension was seeded on the surface of the plate containing Mueller-Hinton agar. Using light pressure, antimicrobial disks were placed on the surface.

Antimicrobial agents were selected according to the identified microorganism based on CLSI recommendations.¹² After an incubation period of 16 to 18 hours at 37°C, the diameter of inhibition halos and the diameter of the disk were measured. The halos were measured in millimeters using a ruler placed against the back of an inverted Petri dish. The area of inhibition was defined as the area without growth detectable by the naked eye. The sizes of growth inhibition halos were interpreted according to CLSI guidance,¹² and the samples were classified as sensitive, intermediate, or resistant to the agents tested.

The open-source Jamovi software (version 1.2.27; https://www.jamovi.org/download.html) was used to analyze the data. Because most of the variables are of a nominal qualitative nature, data non-normality was assumed. Descriptive statistics were performed (percentage and absolute frequency). Inferential statistics were performed using the McNemar test to analyze the hypothesis.

The current study was approved by the ethics committee of the institution in which the study was undertaken under opinion No. 3,443,800, thus meeting all the guidelines and regulatory standards described in Brazilian Resolution No. 466/2012 of the National Health Council. All ethical principles of the Declaration of Helsinki were respected.

The study included 10 males (71.4%) and 4 females (28.6%); the average age was 54.5 years ± 15.2 SD. All patients had preserved physical mobility (without lameness) and level of consciousness. For all patients, nutritional status was adequate and body mass index was normal. Four patients (28.6%) had cancer, and 2 patients (14.3%) had chronic kidney disease; 8 patients (57.1%) had no comorbidities.

In the 10-month study period, 14 PIs were examined in the clinical wards of the treating hospital. No patient had an allergic reaction to PHMB. Table 1 shows PIs occurred predominantly in the sacral area (64.3%), followed by the calcaneal and malleolar areas (14.3% each) and the trochanteric area (7.1%). Nine PIs (64.3%) were stage 3, and 5 (35.7%) were stage 4.

Of note, all 14 patients in the study used medication with a systemic effect, which can interfere with the wound healing process. Of these medications, 64.3% were analgesic and 21.4% were immunosuppressive (eg, corticosteroids).

In 6 lesions (42.9%), granulation tissue covered more than 50% of the wound; in 8 lesions (57.1%), granulation tissue covered 50% or less of the wound. Epithelialization tissue was present on the edges or in dispersed islets in all 14 lesions, and granulation tissue covered 1% to 25% of the wound in 11 lesions (78.6%). Slough, which is a type of necrotic tissue, was present in 12 lesions (85.7%), and 5 lesions (35.7%) had slough in the whole wound bed. Coagulation necrosis was absent in 9 lesions (64.3%). Coagulation necrosis was present in 5 lesions (35.7%), covering 1% to 25% of the wound in 3 lesions (21.4%) and covering 50% or more of the wound in 2 lesions (14.3%).

Ten lesions (71.4%) had serosanguineous exudate, and 4 lesions (28.6%) had serous exudate. The quantity of exudate over 24 hours was measured. Six lesions (42.9%) had exudate covering 25% of the gauze (small amount), 5 lesions (35.7%) had exudate covering half of the gauze (moderate amount, 50%–75%), and 3 lesions (21.4%) had exudate covering 75% or more of the gauze (large amount).

Odor was also evaluated. Four lesions (28.6%) had no odor, and 10 lesions (71.4%) had a foul odor. Reduced intensity of foul odor in these 10 lesions was noted 10 hours after application of PHMB.

Six patients (42.9%) reported pain intensity between 2 and 4 on the VAS, 5 patients (35.7%) reported pain less than 2, and 3 patients (21.4%) reported pain between 4 and 7.

The findings of microbial identification (different types of bacteria) before and after PHMB application are shown in Table 2. In total, 28 wound swab samples were collected—14 before application of PHMB and 14 after application of PHMB. Thirteen different types of bacteria were identified, and in each lesion at least 1 type of bacteria was present prior to application of PHMB. After application of PHMB, only 1 lesion (7.1%) had no microorganism; 21.3% had 1 type of bacteria, 57.1% had 2 types of bacteria, and 14.3% had 3 types of bacteria.

S aureus was the most prevalent microorganism both before and after application of PHMB (swab 1, 21.73%; swab 2, 23.81%), followed by P aeruginosa (swab 1, 17.39%; swab 2, 19.05%); both are aerobic bacteria. After applying PHMB and collecting a second wound swab sample, 21 bacterial species that had been identified in the initial swab sample were still present. Polyhexamethylene biguanide was effective only against Serratia marcescens.

No statistically significant difference was identified concerning qualitative microbiological identification after using PHMB (P =.03). That is, there was no change in the qualitative presence of the microorganisms identified before and after the application of PHMB.

After PHMB was applied, papain was used for enzymatic coverage in 9 lesions (64.3%); in 7 of these lesions, a concentration of 15% papain was used, 3 in association with hydrogel and 4 activated with 10% urea cream. The remaining 2 of 9 lesions were covered with 30% papain activated with 10% urea. In 1 lesion, 15% papain was linked with 10% urea to the activated carbon plate. Polymeric foam with silver was used in 3 lesions (21.4%) and collagenase with chloramphenicol in 1 lesion (7.1%).

Klebsiella pneumoniae showed 100% resistance to the tested antibiotics, Acinetobacter showed 90% resistance, P aeruginosa 88.9%, Citrobacter freundii 87.5%, S aureus 66.7%, and MRSA 57.1%. Only S marcescens showed no resistance to any antibiotic tested. In addition, MRSA was identified in lesion 14.

The results of the current study are similar to studies conducted in Brazil^13-15 and Spain¹⁶ in which individuals with PI were 50 years or older and had comorbidities. Those studies reported a higher prevalence of ulcers in females than in males; in the current study, however, most of the sample population was male (71.4%).

In the study reported herein, most PIs affected the sacral region (64.3%), which is similar to the findings reported by Campanili et al.⁵ Other factors that contribute to the occurrence of PI are age, nutritional status, tissue perfusion, hydration, physical mobility, level of consciousness, comorbidities, friction, shear, humidity, and pressure.² Worldwide, stage 2 PI is the most common.^17-19

Other factors that alter wound healing are prolonged use of medications with a systemic effect, including corticosteroids and antibiotics. Corticosteroids alter the tissue repair process, thereby compromising the resistance of new tissue, and antibiotics interfere in the microbiota of the lesion, resulting in formation of multidrug-resistant bacteria and delayed wound healing or possibly death, which can interfere with and delay wound healing.^20-22 Another complication resulting from indiscriminate use of antibiotics is bloodstream infection, which can cause sepsis.²³ Because the lesions were not followed and prior patient records were not examined to track the status of the lesions over time, it is not possible to determine whether use of antibiotics and corticosteroids altered the healing process.

Lesions were clinically evaluated for signs of infection, especially pain, heat, erythema, edema, and purulent exudate. Other clinical manifestations of infection are foul odor, tunneling to soft tissues, and large quantities of bloody and serous exudate.^7,23

Pain, the fifth vital sign, must be managed correctly; often, this requires the use of pharmacologic measures. The prescription of analgesia is widely accepted, and analgesia is used inside and outside the hospital setting.²⁴ In this study, 9 patients (64.3%) had been treated during the study with analgesic drugs, and 3 reported pain of an intensity between 4 and 7 on the VAS scale over a 24-hour period.

Complications of PI include morbidity and infection, resulting in more extended hospital stays and treatment costs, as well as mortality.¹⁹ In the current study, 5 lesions (21.74%) were colonized by S aureus and 4 lesions (17.39%) by P aeruginosa; both microorganisms are commonly present in PIs.^6,7

According to the literature, PHMB also removes biofilms, acts on resistant bacteria, and promotes reepithelization.^{8,19,20,23,25-27} Wounds infected by bacteria in the sessile form often heal slowly and produce resistant strains of bacteria.^23,27 Polyhexamethylene biguanide is a broad-spectrum synthetic antiseptic; its antimicrobial peptides bind to the bacterial wall, allowing water to enter, resulting in bacterial lysis.²⁰ In the current study, PHMB was shown to have a bactericidal effect on S marcescens alone. However, it contributed to reducing the foul odor in the lesions reevaluated after 24 hours of application. Per the manufacturer,²⁸ in the PHMB instructions, the antimicrobial action of PHMB starts soon after its application.

Microorganisms can delay wound healing, during which time a wound can be colonized by several bacterial strains.²² The current study showed that 13 lesions (92.9%) were colonized with 1 or more bacterial strains. The bacteria are found in the environment as single free cells (ie, planktonic bacteria) or as associated cells anchored to a surface (ie, sessile bacteria). Planktonic bacteria demonstrate greater virulence and greater proliferation, whereas sessile bacteria are more resistant to the environment.^27,29,30

The sessile form of bacteria, also known as biofilm, begins by anchoring the planktonic bacteria to a surface (ie, inorganic matter). This anchoring is done by an extracellular polymeric substance produced by the bacteria itself. This same substance supports the biofilm structure, which can be flat or mushroom-shaped.^27,29,30

Bacterial resistance in lesions can result from the indiscriminate long-term use of antibiotics, and studies show that the sample is resistant to at least 1 type of antibiotic.^7,31,32 Similarly, the current study found that 91.7% of samples were resistant to 1 or more antimicrobial types. The sample of K pneumoniae demonstrated 100% resistance to the tested antibiotics, and only S marcescens was sensitive to all antimicrobial agents and to PHMB.

In the current study, the limitations include the cross-sectional approach, the small number of cases, and the analysis performed in only 2 analyses. A randomized clinical trial is necessary to compare PHMB with other technologies.

The microorganisms present in the PIs at both the first and second wound swab were Acinetobacter spp, C freundii, Enterobacter spp, Enterococcus spp, K pneumoniae, Morganella morganii, P mirabilis, Providencia stuartii, P aeruginosa, S aureus, and MRSA. S marcescens was not detected in the second collection.

K pneumoniae showed 100% resistance to the tested antibiotics, Acinetobacter showed 90% resistance, P aeruginosa 88.9%, C freundii 87.5%, S aureus 66.7%, and MRSA 57.1%. Only S marcescens did not show resistance to any antibiotic tested.

The study findings showed PHMB did not have an effect against the microorganism tests, only S marcescens. Studies with larger populations and randomized clinical trial methodologies should be conducted to elucidate other findings related to PHMB.

Authors: Felipe Monteiro Vasconcelos, RN¹; Carolina Cabral Pereira da Costa, PhD¹; Ellen Marcia Peres, PhD¹; Robson Souza Leão, PhD¹; Helena Ferraz Gomes, PhD¹; Priscila Cristina da Silva Thiengo Andrade, PhD¹; Ariane da Silva Pires, PhD¹; Cristiene Faria, RN²; Ana Paula de Oliveira Motta, RN²; and Bruna Maiara Ferreira Barreto Pires, PhD³

Affiliations: ¹Universidade do Estado do Rio de Janeiro, Rio de Janeiro, Brazil; ²Hospital Universitario Pedro Ernesto, Rio de Janeiro, Brazil; ³Universidade Federal Fluminense, Rio de Janeiro, Brazil

Disclosure: The authors disclose no financial or other conflicts of interest.

ORCID: Felipe Monteiro Vasconcelos, 0000-0002-0462-5935; Carolina Cabral Pereira da Costa, 0000-0002-0365-7580; Ellen Marcia Peres, 0000-0003-4262-6987; Robson Souza Leão, 0000-0003-0636-1520; Helena Ferraz Gomes, 0000-0001-6089-6361; Priscila Cristina da Silva Thiengo de Andrade, 0000-0003-0840-4838; Ariane da Silva Pires, 0000-0003-1123-493X; Cristiene Faria, 0000-0001-6548-1851; Ana Paula de Oliveira Motta, 0000-0003-2728-6697; Bruna Maiara Ferreira Barreto Pires, 0000-0002-5584-8194

Correspondence: Bruna Maiara Ferreira Barreto Pires, PhD, Professor, Universidade Federal Fluminense, Fundamentos de Enfermagem, Rua Dr Celestino, 74, Centro, Rio de Janeiro, Brazil 20550-900, Brazil; bruna.barreto07@gmail.com