Application of bacterial cellulose film as a wound dressing in varicose vein surgery: A randomized clinical trial

Abstract

Background

Chronic venous insufficiency is common all over the world. It is often related to lower limb varicose veins. Surgical treatment is necessary in most cases but the management of the operative wounds remains a challenge. Nowadays, there is no consensus on which is the ideal dressing for covering operative wounds, but features such as low cost, hypoallergenicity, and good adherence are expected. In this study, a material that meets all these features, the bacterial cellulose film (BCF), was evaluated in patients who underwent varicose vein surgery.

Methods

A randomized, prospective clinical trial was carried out at the Angiology and Vascular Surgery Department of the Hospital das Clínicas/EBSERH - UFPE, from May 2023 to September 2024. The study included 55 patients who underwent varicose vein surgery, divided in two groups: an experimental group, which used a BCF to cover microincisions, and a control group. which used microporous tape. Between postoperative days 4 and 6, patients were assessed for pain, itching, and wound appearance using the Southampton Wound Assessment Scale. The groups were homogeneous in all demographic data, presence of comorbidities, and Clinical-Etiology-Anatomy-Pathophysiology Clinical classification.

Results

Pain on removal was more common in the control group with a mean scale score of 3.46 ± 3.26 compared with 0.63 ± 1.38 (P < .001).

Conclusions

There was no statistical difference between the groups in itching and wound healing. BCF showed healing results comparable with microporous tape, with the advantage of generating less pain on removal.

Article Highlights.

• •Type of Research: Randomized controlled trial
• •Key Findings: Bacterial cellulose film is a safe, effective dressing for varicose vein surgery, promoting rapid operative wound healing, with wounds closing in approximately 3 days. It also causes less pain during application and removal than microporous tape.
• •Take Home Message: This study pioneers the use of bacterial cellulose film in operative wounds after varicose vein removal. Further research, including larger multicenter trials, is needed to confirm and expand these findings.

Conventional surgery for lower limb (LL) varicose veins involves removing dilated veins through micro-incisions. These small operative wounds (OWs) are usually not sutured to avoid scarring and are instead covered with microporous tape. Although intended to protect the incision and aid healing, the strong adhesiveness of microporous tape can make handling and removal difficult, often leading to local skin reactions such as dermatitis, folliculitis, maceration, and mechanical trauma—classified as medical adhesive-related skin injuries). In the United States, medical adhesive-related skin injury treatment costs exceed $11 million annually, yet evidence supporting the effectiveness of the microporous tape in varicose vein surgery remains limited.1, 2, 3, 4

Considering the lack of standard associated with OW dressings in varicose vein treatment, the development of state-of-the-art dressings and the evaluation of innovative biomaterials and medical products are highly relevant. Such advancements can accelerate healing, reduce patient discomfort, and enhance the efficiency of care provided by nursing and multidisciplinary teams, while also decreasing hospital stays and overall health care costs.^5,6

This study proposes a biopolymer dressing made from bacterial cellulose obtained from molasses of sugar cane, known as bacterial cellulose film (BCF), as an alternative to microporous tape in LL varicose vein surgery OWs. In previous studies, the BCF was demonstrated to be a safe, biocompatible, and low-toxicity material with physicochemical properties that promote healing.^7,8

Its semitransparent and naturally adherent nature allows wound monitoring without removal, while protecting against dermatitis and pain. Derived from renewable sources, BCF stands out from synthetic membranes like polyurethane. It has shown promising results in preclinical models and various medical applications in humans, including arterial and bladder reconstruction,^9,10 treatment of venous and arterial ulcers,^11,12 and surgical wound care in hypospadias.¹³ Although effective in other contexts, its use in varicose vein surgery OWs—characterized by specific dressing needs, such as to avoid bruising—had not yet been explored, despite its demonstrated stability, exudate-responsive component release, and suitability for less frequent dressing changes.^11,14, 15, 16

Finally, this randomized clinical study aims to evaluate the effectiveness of BCF as a dressing for LL varicose vein surgical wounds, compared with conventional microporous tape. The study focuses on key outcomes including wound healing progression, incidence of pruritus, pain during dressing removal, and overall patient comfort.

Methods

A randomized, prospective, single-blind clinical trial was conducted at the Department of Angiology and Vascular Surgery of the Hospital das Clínicas – Empresa Brasileira de Serviços Hospitalares - Universidade Federal de Pernambuco (HC/EBSERH-UFPE), between May 2023 and September 2024. The study was approved by the Human Research Ethics Committee (approval number: 5,855,571) and included 55 patients with LL varicose veins who underwent surgical treatment. All participants signed the Free and Informed Consent Form. After initial data collection, patients underwent phlebectomy through mini-incisions and were randomly allocated into two groups. Randomization was performed using the Randomizer program (available at randomizer.org). Patients in the experimental group (EG) received BCF for microincision coverage, whereas those in the control group (CG) were treated with microporous tape. The inclusion criteria encompassed patients diagnosed with LL varicose veins who had undergone surgical treatment. Exclusion criteria included failure to attend the scheduled postoperative follow-up appointment within 4 to 6 days after surgery.

Sample size calculation was conducted using a free online tool provided by the Massachusetts General Hospital Mallinckrodt General Clinical Research Center, accessible at http://hedwig.mgh.harvard.edu/sample_size/size.html. The parameters considered included a significant level of 95% (P < .05), statistical power of 80%, and a standard deviation of 2.57 for the mean individual difference on the visual analog scale (VAS) for pain. A minimum important difference of 2.0 points on the VAS was considered clinically relevant. Based on these criteria, the estimated required sample size was 54 participants.

The application of dressings was performed in accordance with a standardized protocol for all EGs. All OWs were cleansed meticulously using compresses and saline solution before the application of dressings. In the EG, BCF dressings were premoistened with saline to activate their adhesive properties before application to the skin. Subsequently, all patients received compresses and bandages, followed by the application of 7/8 compression stockings with 20 to 30 mm Hg pressure. Patients were blinded to the type of dressing used owing to its placement beneath compresses, bandages, and elastic stockings. The primary outcome of interest was the assessment of surgical wounds within 5 days after the procedure.

All patients were instructed not to remove their compression stockings or wet their legs until the follow-up appointment, scheduled between 4 and 6 days after surgery. At that time, with the patient in the supine position—preventing them from seeing which type of dressing had been applied—the compression stockings, bandages, and compresses were removed carefully. Patients were then asked to report any pain experienced during dressing removal and to rate it using a VAS ranging from 0 (no pain) to 10 (worst possible pain). Pain was recorded immediately after dressing removal. Additionally, patients were asked whether they had experienced any itching while using the dressing.

After this evaluation, standardized photographic documentation of the surgical sites on the LL was performed using a 12-megapixel regular camera system, capable of capturing high-resolution images at 4032 × 3024 (iPhone X). Images were captured from the following anatomical regions: right anterior thigh, right posterior thigh, left anterior thigh, left posterior thigh, right anterior leg, right posterior leg, left anterior leg, and left posterior leg. Afterward, an independent vascular surgeon assessed the OWs in each area, using the photographic record, according to the Southampton Wound Assessment Scale (SWAS).¹⁷ The worst pattern found in the area was considered.

Statistical analyses were conducted using IBM SPSS Statistics version 20.0 and SigmaPlot version 12.0 (Systat Software). Data were presented as mean ± standard deviation, mean difference with 95% confidence intervals, or as absolute numbers with corresponding percentages.

Categorical variables were compared between groups using the χ² test or Fisher exact test, as appropriate. The Shapiro-Wilk test was used to assess the normality of numerical variables. For variables with a parametric distribution, comparisons between groups were made using the Student t test. For nonparametric data, the Mann-Whitney U test was applied. A P value of less than .05 was considered statistically significant.

This study was registered in the Brazilian Registry of Clinical Trials (ReBEC) under the number: RBR-99f794z.

Results

Sixty-five patients were assessed for eligibility and were randomized for allocation to the CG or the EG (Fig 1). After surgery, during follow-up, nine patients did not return for evaluation: six from the CG and three from the EG. In addition, one patient from the CG was excluded for removing the dressings just 2 days after surgery (discomfort with the use of elastic stockings). Thus, 55 patients remained fit for analysis, 27 patients in the CG and 28 in the EG, corresponding with 393 operated areas (208 from the CG and 185 from the EG).

Fig 1.

Flow of participants in each group. Created with BioRender.com.

The sociodemographic profile of the patients was evaluated, including age, gender, race, health conditions and medical history, as well as level of physical activity. Among the 55 participants assessed (Table I), the average age was 50.6 ± 9.7 years, with a predominance of females (89%). According to Fitzpatrick scale,¹⁸ 27 patients were categorized as skin type IV (49.1%), 16 type II (29.1%) and 12 type V (21.8%); 2 declared themselves smokers (3.6%), 49 patients had thrombosis already (89%), 33 had had a previous surgery (60%), 38 had a family history of varicose veins (69%), and 25 practiced regular physical activity (45.4%). No statistical difference was observed between the variables analyzed (P > .05) (Table I).

Table I.

Sociodemographic and clinical characteristics of the study participants (n = 55)

Variable	Value
Mean age, years	50.6 ± 9.7
Gender
Female	49 (89.0)
Male	6 (11.0)
Race/ethnicity
Brown	27 (49.1)
White	16 (29.1)
Black	12 (21.8)
Smokers	2 (3.6)
History of thrombosis	49 (89.0)
Previous surgery	33 (60.0)
Family history of varicose veins	38 (69.0)
Regular physical activity	25 (45.4)

Values are mean ± standard deviation or number (%).

Table II shows the Clinical-Etiology-Anatomy-Pathophysiology Clinical classifications,¹⁸ the most frequent of which were C2 and C3, both in the CG and the EG. The difference between the groups was not statistically significant (P = .658).

Table II.

Clinical-Etiology-Anatomy-Pathophysiology (CEAP) clinical classification in the groups studied

CEAP classification	CG (n = 27)	EG (n = 28)	P valuea
C1	0 (0)	0 (0)	.658
C2	12 (44)	9 (32)
C3	12 (44)	17 (61)
C4	2 (7)	1 (4)
C5	1 (4)	1 (3)

CG, Control group; EG, experimental group.

Values are number (%).

^aComparison between CG and EG; χ² or Fisher's exact test were used for categorical variables.

Crossectomy and/or saphenectomy were the same between the groups (P > .05), with 19 and 15 in the CG and 17 and 11 in the EG (Table III). Dressings were applied immediately after surgery (Fig 2).

Table III.

Crossectomy/saphenectomy performed during varicose vein surgery in the sample

Variables	CG (n = 27)	EG (n = 28)	P valuea
Saphenectomy
Yes	15 (56)	11 (39)	.348
No	12 (44)	17 (61)
Crossectomy
Yes	19 (70)	17 (61)	.639
No	8 (30)	11 (39)

CG, Control group; EG, experimental group.

Values are number (%).

^aComparison between the CG and EG; χ² test or Fisher's exact test were used for categorical variables.

Fig 2.

Application of dressings in the immediate postoperative period. (A) Experimental group and (B) control group (CG).

During the postoperative assessment, pain during dressing removal was reported by 19 patients (70%) in the CG who received microporous tape, compared with only 5 patients (18%) in the EG who received BCF (Table IV). The mean pain score on VAS was 3.27 in the CG and 0.61 in the EG, with a mean difference of 2.66 (95% confidence interval, 1.34-3.98). This difference was statistically significant, indicating significantly less pain in the EG (P < .001) (Table IV). Pruritus was reported by 16 patients (59%) in the CG and 13 patients (46%) in the EG; however, this difference was not statistically significant (P = .495) (Table IV).

Table IV.

Assessment of pain and itching

Outcome	CG (n = 27)	EG (n = 28)	OR (95% CI)	Diff mean (95% CI)	P value
No. of areas assessed/patient	6.41 ± 1.74	6.50 ± 1.62	NA	−0.09 (−1.00 to 0.82)	.887
Pain	3.27 ± 3.16	0.61 ± 1.40	NA	2.66 (1.34 to 3.98)	<.001a
Pain when removing dressing			10.93 (3.06 to 38.98)
Yes	19 (70)	5 (18)		NA	<.001a
No	8 (30)	23 (82)		NA
Itching			1.68 (0.58 to 4.88)
Yes	16 (59)	13 (46)		NA	.495
No	11 (41)	15 (56)		NA

CG, Control group; EG, experimental group; NA, not available.

Data expressed as mean ± standard deviation or number (%).

^aComparison between the CG vs the EG; Student t test or Mann-Whitney U test were used for continuous variables and χ² test or exact test were used for categorical variables.

Regarding the assessment of LL areas using the SWAS, 62 areas (30%) in the CG demonstrated normal healing, compared with 64 areas (35%) in the SG. Additionally, 101 areas (49%) in the CG presented with mild ecchymosis (grade Ia), and 38 areas (18%) with more pronounced ecchymosis (grade Ib), whereas in the SG, 85 areas (46%) showed grade Ia and 32 areas (17%) grade Ib ecchymosis. Localized erythema (grade IIa) was observed in seven areas (3%) in the CG and in three areas (2%) in the SG. No statistically significant differences were found between the groups. In both groups, there were no cases of infection and there was no case of interruption of the research because of pain or pruritus. Detailed wound assessment results are presented in Table V.

Table V.

Wound assessment by Southampton Wound Assessment Scale (SWAS)

SWAS	GC (n = 208)	GE (n = 185)	P valuea
0- Normal	62 (30)	64 (35)	.570
Ia- Some ecchymosis	101 (49)	86 (46)
Ib- Considerable ecchymosis	38 (18)	32 (17)
Ic- Medium erythema	-	-
IIa- Erythema and other signs of inflammation at one point	7 (3)	3 (2)
IIb- Erythema and other signs of inflammation around the suture	-	-
IIc- Erythema and other signs of inflammation along the wound	-	-
IId- Erythema and other signs of inflammation around the wound	-	-
IIIa- Serous/hematous drainage at one point (<2 cm)	-	-
IIIb- Serous/hematous drainage along the wound	-	-
IIIc- Large volume serous/chemo drainage	-	-
IIId- Prolonged serous/chemo drainage (>3 days)	-	-
IVa- Pus at one point	-	-
IVb- Pus along the wound	-	-

CG, Control group; EG, experimental group.

Data expressed as number (%).

^aComparison between group 1 vs group 2 for evaluator 2; 1 comparison between group 1 vs group 2 for evaluator 3; χ² test or Fisher's exact test were used for categorical variables.

Discussion

The search for wound dressings that combine comfort, safety, and effectiveness in the healing process has been a central topic in the management of surgical and chronic wounds, focusing on improving patient quality of life and minimizing associated complications.^19,20 In this study, the BCF demonstrated efficacy in treating surgical wounds after varicose vein surgery in the LL, with a significant reduction in pain during dressing removal compared with conventional microporous tape (P < .5).

The strategy of analyzing several areas in the same patient was chosen because the same patient could have different results in different areas. The same procedure was undertaken by Fonseca et al.⁴ Microporous tape was used in the CG because, in Brazil, it is the one of the most used dressings for varicose vein surgery. Because BCF is an experimental dressing, only estimated market values are available; therefore, its financial cost was not evaluated.

Although both dressings showed comparable outcomes in terms of wound healing, BCF provided greater patient comfort and reduced cutaneous trauma during dressing changes, corroborating previous findings on the favorable properties of bacterial cellulose for wounds of different etiologies.²¹ For example, Cavalcanti et al²² reported that patients treated with bacterial cellulose membrane dressings experienced less pain during removal and were able to maintain the dressing for up to 7 days, whereas conventional dressings required changes every 2 days. Similarly, Silva et al¹¹ observed faster initial wound area reduction in patients with venous ulcers treated with the biopolymer dressing compared with rayon gauze, despite fewer dressing changes and the ability to leave the membrane in place even during bathing.

The pain experienced during removal of conventional microporous tape dressings may be attributed to their strong adhesion to the wound bed, leading to secondary injuries and discomfort, especially in areas with more delicate skin.^3,23,24 This characteristic also likely explains the higher frequency of erythema observed in the CG, although without statistical significance (P > .5), as the aggressive adhesion mechanism of microporous tape compromises the skin barrier and triggers an inflammatory response.²⁴ In contrast, BCF presents moderate adhesion and relies on its hydrophilic properties, facilitating removal without causing additional pain or tissue damage.^11,25

It is worth noting that the documented skin abrasions observed in the CG highlight the potential for secondary trauma and discomfort associated with the high adherence of microporous tape. In contrast, the BCF dressing after 5 days of application showed moderate adhesion and hydrogel-like properties, which support atraumatic removal and greater patient comfort. These findings underscore the importance of selecting wound dressings that minimize trauma during removal, thereby reducing complications such as skin abrasions, inflammation, and patient discomfort.^1,25 In the postoperative setting of varicose vein surgery, inadequate dressing adaptation or overly frequent changes can contribute to significant local complications, including skin abrasions, inflammatory reactions, and infections, ultimately delaying healing.²⁶

The use of the SWAS was appropriate, given the absence of a specific instrument for postvaricose vein surgery wounds. This descriptive scale allowed for standardized and reproducible analysis of healing outcomes, aligning with the methodology proposed by Bailey et al.¹⁷ The absence of infections or severe complications in both groups reinforces the safety of BCF as a postoperative wound covering. Furthermore, a significant advantage of BCF lies in its production from renewable sources, rendering it a sustainable and economically viable material for clinical applications.⁸ Its compliance with the biocompatibility and safety criteria established by the US Food and Drug Administration further substantiates its suitability as a wound dressing.¹⁶

The main limitation of this study was the sample, which consisted primarily of patients classified as Clinical-Etiology-Anatomy-Pathophysiology C2 and C3, restricting the evaluation of the BCF in individuals with more advanced skin trophic changes. This limitation suggests that caution should be exercised when generalizing these findings to patients with more severe forms of chronic venous insufficiency. To address these gaps, multicenter studies with a broader range of clinical profiles and diverse risk characteristics are warranted. Such studies would not only help to confirm and refine the current results, but also explore the performance of BCF in other surgical applications and in populations at higher risk for wound healing complications. The small number of patients enrolled in this research could affect its power.

Previous studies have scientifically analyzed and characterized the physical and chemical properties of BCF, confirming it as a suitable candidate for applications in regenerative medicine by promoting the re-epithelialization of surgical wounds, thereby supporting the findings of the present study. These findings suggest that the use of BCF may represent a promising option for the management of surgical wounds, offering increased patient comfort, benefiting clinical practice, and contributing to a more efficient and less painful recovery after varicose vein procedures.

However, because this is a pioneering study on the use of BCF in OWs resulting from varicose vein removal, further complementary studies are necessary. These studies should include multicenter clinical trials and a larger sample size to broaden the applicability of the dressing and validate the results obtained.

Conclusions

Based on the results obtained, it was concluded that BCF can be used effectively and safely as a dressing in varicose vein surgery, supporting the healing process of small OWs, with complete wound closure occurring on average within 3 days. Furthermore, BCF presents the advantage of causing less pain during dressing application and removal when compared with microporous tape.

Author contributions

Conception and design: AM, FP, JA

Analysis and interpretation: EL, FR, LM, SP, KS, SP

Data collection: AM, FR, TP, MN, JB, JS

Writing the article: AM, LM, TP, MN, SP, KS, JB, SP, JS

Critical revision of the article: AM, EL, FR, LM, FP, TP, MN, SP, KS, JB, SP, JS, JA

Final approval of the article: AM, EL, FR, LM, FP, TP, MN, SP, KS, JB, SP, JS, JA

Statistical analysis: Not applicable

Obtained funding: Not applicable

Overall responsibility: EL

Funding

Partially funded by Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq), Brazil: J.L.A.A [CNPq/MCTI 424,123/2021-5]. e Fundação de Amparo à Ciência e Tecnologia do Estado de Pernambuco (FACEPE), Brazil: L.C.M. [FACEPE BFP-0093-4.01/24].

Disclosures

None.