Risk factors for medical adhesive-related skin injury at the site of peripherally inserted central venous catheter placement in patients with cancer: a single-centre prospective study from China

Abstract

Objectives

This study aims to explore the incidence of, and risk factors for medical adhesive-related skin injury (MARSI) at peripherally inserted central venous catheter (PICC) sites in patients with cancer.

Design

A prospective observational cohort study was conducted at a tertiary hospital in Shenzhen, China.

Setting

This was a single-centre study conducted in a tertiary hospital in Shenzhen, China.

Participants

A total of 340 patients with cancer and PICC placement from January 2022 to June 2023 were selected using a convenience sampling method.

Methods

Factors potentially associated with PICC-related MARSI (PICC-MARSI) were recorded, including patient demographics, and catheter placement and maintenance. Patients were divided into MARSI and non-MARSI groups. Univariate analysis was performed to screen for associated variables, and logistic regression analysis was used to identify independent risk factors for PICC-MARSI.

Results

Of all 340 patients enrolled, 33 (9.7%) developed PICC-MARSI, including skin tear (8, 24.2%), tension injury (5, 15.2%), irritant contact dermatitis (10, 30.3%), allergic dermatitis (7, 21.2%) and maceration (3, 9.1%). Multivariable analysis showed that age (OR=1.058, p=0.001, 95% CI 1.023–1.094), wet skin (OR=4.873, p=0.003, 95% CI 1.728–13.742), dry skin (OR=6.247, p<0.0001, 95% CI 2.239–17.431), oedema (OR=3.302, p=0.008, 95% CI 1.365–7.985), allergy history (OR=6.044, p=0.001, 95% CI 2.040–17.906), dressing type (OR=3.827, p=0.003, 95% CI 1.595–9.185), body mass index (BMI) <18.5 (OR=4.271, p=0.015, 95% CI 1.327–13.742) and BMI 25–30 (OR=2.946, p=0.027, 95% CI 1.131–7.678) were independent risk factors for PICC-MARSI.

Conclusions

Proper catheter maintenance and appropriate dressing selection are crucial for the prevention of this condition.

Strengths and limitations of this study.

• This is a prospective cohort study on the incidence of, and risk factors for medical adhesive-related skin injury (MARSI) at peripherally inserted central venous catheter (PICC) sites in patients with cancer.

• This study analyses the risk factors for PICC-related MARSI and describes the procedural details and precautions taken by our team to prevent MARSI in our centre.

• To some extent, the convenience sampling used in this study weakens the results.

• The same catheter material and maintenance method were used, and all the patients with PICC were patients that required chemotherapy, which may have led to other risk factors being missed.

Introduction

Medical adhesive-related skin injury (MARSI) is defined as skin damage that occurs after removing medical adhesive products, with symptoms including erythema lasting for ≥30 min, accompanied by blisters, erosion or tearing. MARSI includes skin stripping, tension injury or blister, skin tear, irritant contact dermatitis, allergic dermatitis, maceration and folliculitis.¹ In 2017, the World Congress of Vascular Access defined central vascular access device (CVAD)-associated skin impairment (CASI) based on the definition of MARSI.² A study of 1172 patients with cancer and CVAD placement found that 24.1% of CVAD sites had MARSI, and those with peripherally inserted central venous catheters (PICC) placement sites had a higher incidence of MARSI than those with other types of CVADs.³

PICCs can be left in patient blood vessels for up to a year and are widely used in long-term cancer treatment; however, repeated application and removal of adhesive products during catheter maintenance can increase the risk of MARSI, particularly in patients with cancer who have skin, metabolism or immune dysfunction.⁴ This can cause local itching, pain, and increase the risk of catheter infection and unplanned removal^{5 6}; however, skin damage is often overlooked in studies of PICC-related complications, and there is a lack of relevant research data.^{7 8} Therefore, the aim of this study was to explore the risk factors for PICC-related MARSI (PICC-MARSI) in patient with cancer, to provide a reference for the prevention and management of this condition.

Methods

Design

This single-centre prospective observational cohort study was conducted at a tertiary hospital in Shenzhen, China. This study followed the Strengthening the Reporting of Observational Studies in Epidemiology statement for reporting of observational studies in epidemiology (online supplemental file 1).

Setting

This study was conducted in a tertiary care hospital and the departments involved in the work included oncology, haematology, gastroenterology, respiratory medicine, thoracic surgery, breast surgery and the intensive care unit.

Patients required PICC placement and chemotherapy at a tertiary hospital in Guangdong Province from January 2022 to June 2023 were selected using a convenience sampling method. Patients used 4Fr single-lumen triple-valve PICC catheters produced by Bard, USA, and disposable catheter maintenance kits. Dressings included 10×12 cm Tegaderm HP transparent film from 3M Health Company and 10×12 cm IV3000 from Smith & Nephew, London, UK.

Study participants

This study included patients with cancer who met the following criteria: clinical diagnosis of cancer, age ≥18 years, PICC placement with available catheter placement information, ability to communicate with investigators and voluntary participation with signed informed consent. The exclusion criteria were: congenital or past skin diseases, skin damage, allergies or dermatitis at the time of PICC placement, withdrawal from the study, and abnormal mental behaviour.

Data collection

The study team reached a consensus on the definition of MARSI, data collection content and methods through meetings. An information collection form was developed to collect PICC-MARSI-related data from patients with cancer based on a literature review of MARSI-related risk factors. Within 24 hours of catheter placement, the placement nurse recorded patient characteristics, placement information and laboratory tests, including: age, gender, consciousness, body mass index (BMI), tumour type, comorbidities, Barthel index of Activities of Daily Living score, smoking, allergy history, CVAD history, needle chemotherapy history, patient compliance, puncture vessel, catheter side limb, lidocaine wet compress, modified skin incision, smooth puncture process and albumin level. Catheter maintenance was performed by nurses with PICC maintenance qualifications (PICC maintenance specialist certificate) every 7 days. Nurses observed skin problems when removing the medical adhesive and recorded the skin condition at the catheter placement site (normal, dry, wet, oedema), skin antisepsis, use of glucocorticoids and adhesive product use. The study team evaluated whether MARSI occurred and the type of damage after 30 min, using the International Skin and Stoma Care Association’s 2013 expert consensus criteria. If MARSI occurred, data collection was stopped. The completed survey forms were collected by the researcher and medication information was obtained by reviewing the Hospital Information System. Data were double-checked and entered into Excel 2013 to establish a database.

Sample size calculation

According to the calculation formula for estimating population rates through sampling surveys: n=Z² _α/2 P(1–P)/δ², assuming an error allowance of δ=0.05, and a MARSI incidence rate of approximately 3.4%–29.89% (mean, 16.6%), α=0.05, Z_0.05/2=1.96 and p=0.166, the calculated sample size was n=213.^{9 10} Considering the exclusion criteria, a dropout rate of 10% was set, and the required sample size was determined to be ≥235.

Data analysis

All data collected in Excel 2013 were imported into SPSS V.24.0 for statistical description and analysis. Count data for patient demographics are described using frequency and composition ratios, and continuous variables are described using mean±SD values. Patients were divided into MARSI and non-MARSI groups, based on PICC placement site. The t-test was used to assess the significance of differences between continuous variables, and the χ² test and Fisher’s exact probability method were used for categorical variables in univariate analysis, with p<0.05 as the significance level. Logistic regression analysis was used to screen for independent risk factors for PICC-MARSI in patients with cancer.

Patient and public involvement

Patients and/or the public were not involved in the design, or conduct, or reporting, or dissemination plans of this research.

Results

A total of 366 patients participated in the study, of which 8 patients withdrew, 13 had missing data and 5 patients died before they could complete the study. A total of 340 cancer patients with PICC placement were finally included in the study, comprising 132 males and 208 females, with mean ages of 59.6±15.0 years and 54.0±15.6 years, respectively. Of these patients, 116 had gynaecological tumours, 69 had haematological tumours, 41 had respiratory system tumours, 80 had digestive system tumours and 34 had other tumours. Thirty-three patients (9.7%) developed PICC-MARSI, including 8 cases (24.2%) of skin tear, 5 cases (15.2%) of tension injury, 10 cases (30.3%) of irritant contact dermatitis, 7 cases (21.2%) of allergic dermatitis and 3 cases (9.1%) of maceration (table 1).

Table 1.

Skin injury types of medical adhesive-related skin injury (MARSI) in this survey

Types of MARSI	n	%
Skin tear	8	24.2
Tension injury	5	15.2
Irritant contact dermatitis	10	30.3
Allergic dermatitis	7	21.2
Maceration	3	9.1

Univariate analysis showed that age (t=–3.213, p=0.001), skin condition (χ²=17.314, p<0.0001), comorbidities (χ²=7.551, p=0.006), oedema (χ²=11.238, p=0.001), BMI (χ²=9.936, p=0.006), allergy history (χ²=6.467, p=0.011) and dressing type (χ²=8.742, p=0.003) were significantly associated with PICC-MARSI (p<0.05) (table 2).

Table 2.

Comparison between participants with and without MARSI at peripherally inserted central venous catheter sites based on personal characteristics and catheter-related characteristics variables

Variable	Non-MARSI group (n=307)		MARSI group (n=33)		t/χ2 test (95% CI)	P value
	n/ $\stackrel{-}{x}$	%/μ	n/ $\stackrel{-}{x}$	%/μ
Age	55.3	15.6	64.4	12.6	−3.213	0.001*
Gender					1.436	0.231
Female	191	91.8	17	8.2
Male	116	87.9	16	12.1
Tumour type					4.123	0.380
Gynaecological	107	92.2	9	7.8
Respiratory	39	95.1	2	4.9
Digestive	68	85.0	12	15.0
Haematological	62	89.9	7	10.1
Other	31	91.2	3	8.8
Skin condition					17.314	0.000*
Normal	212	95.1	11	4.9
Wet	49	83.1	10	16.9
Dry	46	79.3	12	20.7
The Barthel index of ADL					0.058	0.810
61–100	274	90.4	29	9.6
≤60	33	89.2	4	10.8
Comorbidities					7.551	0.006*
None	161	94.7	9	5.3
Present	146	85.9	24	14.1
Oedema					11.238	0.001*
No	246	93.2	18	6.8
Yes	61	80.3	15	19.7
Consciousness					0.111	0.739
Alert	297	90.5	31	9.5
Comatose	10	83.3	2	16.7
BMI					9.963	0.006*
18.5–24.9	185	93.9	12	6.1
<18.5	26	76.5	8	23.5
25–30	96	88.1	13	11.9
Smoking					0.380	0.538
No	255	90.7	26	9.3
Yes	52	88.1	7	11.9
Allergy history					6.467	0.011*
None	275	92.0	24	8.0
Present	32	78.0	9	22.0
CVAD history					0.352	0.553
None	246	90.8	25	9.2
Present	61	88.4	8	11.6
Needle chemotherapy history					0.523	0.469
None	234	89.7	27	10.3
Present	73	92.4	6	7.6
Albumin level					2.231	0.334
Normal	212	91.8	19	8.2
Low	88	87.1	13	12.9
High	7	87.5	1	12.5
Patient compliance					0.272	0.602
Yes	283	90.7	29	9.3
No	24	85.7	4	14.3
Lidocaine wet compress					0.006	0.940
No	95	90.5	10	9.5
Yes	212	90.2	23	9.8
Skin antisepsis					0.022	0.883
Povidone iodine	125	90.6	13	9.4
CHG	182	90.1	20	9.9
Dressing type					8.742	0.003*
IV3000	193	94.1	12	5.9
Tegaderm HP	114	84.4	21	15.6
Puncture vessel					1.095	0.795
Femoral vein	271	90.0	30	10.0
Jugular vein	10	100.0	0	0.0
Subclavian vein	14	93.3	1	6.7
Median cubital vein	12	85.7	2	14.3
Catheter side limb					1.601	0.206
Left	98	93.3	7	6.7
Right	209	88.9	26	11.1
First puncture success					0.001	0.970
No	241	90.3	26	9.7
Yes	66	90.4	7	9.6
First catheterisation success					0.000	1.000
No	35	89.7	4	10.3
Yes	272	90.4	29	9.6
Modified skin incision					1.333	0.248
Yes	257	91.1	25	8.9
No	50	86.2	8	13.8
Use of glucocorticoids					1.052	0.305
No	282	91.0	28	9.0
Yes	25	83.3	5	16.7

Numbers marked with * are significant with a p-value <0.05

ADL, activities of daily living; BMI, body mass index; CHG, chlorhexidine gluconate; CVAD, central vascular access device; MARSI, medical adhesive-related skin injury.

Multivariable analysis showed that age (OR=1.058, p=0.001, 95% CI 1.023–1.094), wet skin (OR=4.873, p=0.003, 95% CI 1.728–13.742), dry skin (OR=6.247, p<0.0001, 95% CI 2.239–17.431), oedema (OR=3.302, p=0.008, 95% CI 1.365–7.985), allergy history (OR=6.044, p=0.001, 95% CI 2.040–17.906), dressing type (OR=3.827, p=0.003, 95% CI 1.595–9.185), BMI<18.5 (OR=4.271, p=0.015, 95% CI 1.327–13.742) and BMI 25–30 (OR=2.946, p=0.027, 95% CI 1.131–7.678) were independent risk factors for PICC-MARSI (table 3).

Table 3.

Results of multivariable analysis of baseline variables associated with peripherally inserted central venous catheter related medical adhesive-related skin injury

Variable	β	SE	Wald	P value	OR	95% CI
Age	0.056	0.017	10.679	0.001	1.058	1.023	1.094
Skin condition			15.032	0.001
Wet	1.584	0.529	8.965	0.003	4.873	1.728	13.742
Dry	1.832	0.524	12.243	0.000	6.247	2.239	17.431
Oedema (yes)	1.194	0.451	7.029	0.008	3.302	1.365	7.985
Allergy history (present)	1.799	0.554	10.540	0.001	6.044	2.040	17.906
Dressing type (Tegaderm HP)	1.342	0.447	9.028	0.003	3.827	1.595	9.185
BMI			7.612	0.022
<18.5	1.452	0.596	5.928	0.015	4.271	1.327	13.742
25–30	1.081	0.489	4.889	0.027	2.946	1.131	7.678
Constant	−8.505	1.411	36.325	0.000	0.000	–

BMI, body mass index.

Discussion

Incidence of PICC-MARSI

The incidence rate of MARSI in non-critical emergency patients has been reported as 3.4%–25.0%.⁹ Konya et al determined that the incidence of skin damage caused by medical adhesive at infusion sites in patients >65 years of age undergoing nutritional support therapy was 20%,¹¹ while the incidence of MARSI in paediatric patients was 14%.¹² In this study, the incidence of MARSI at the PICC placement site in patients with cancer was 9.7%, which is lower than the 29.83% reported by Zhao et al ¹⁰; that study showed that mechanical skin damage is the most common type of MARSI, with pressure damage the most common type of mechanical damage; however, in this study, contact dermatitis (30.3%) and allergic dermatitis (21.2%) were the main types of injury, followed by mechanical damage, including skin tears caused by pressure damage (24.2%) and tension injuries leading to epidermal separation and blister formation (15.2%). In this study, pressure damage to local skin around the PICC mainly manifested as grade I and II pressure damage at the interface of the catheter end interface and the heparin cap interface, similar to the results reported by Noonan et al and Zhao et al.^{10 13} Tension injuries were mainly caused by the improper operation of medical adhesive without tension or by tension blisters caused by patient arm movement and traction of the dressing.

Mechanical skin damage is often easy to prevent or avoid through intervention. The incidence of MARSI in this study was relatively low compared with other reports, mainly reflected in the lower incidence of mechanical damage. Most mechanical damage is caused by improper catheter maintenance operations.^{2 14} All of our PICC maintenance was performed by nurses with professional qualifications (catheter maintenance certificates). The specialist nursing team holds a learning, case discussion and research sharing session every 2 months, to ensure maximal homogenisation of PICC maintenance. Catheter fixation was mainly conducted using transparent dressings and non-sterile tapes. Nurses were required to secure the infusion interface in an elevated position, without direct contact with the skin, and then fix it again with tape. Tension injuries are usually caused by the patient’s arm being kept straight when applying the film. Even if the film is gradually applied without tension from the puncture point to the edge, there is still tension when the elbow moves. Therefore, we adhere the upper part when the patient’s elbow is straight and the lower part when it is bent, to ensure that the skin under the adhesive is tension-free when the patient’s elbow moves. In addition, after identifying the type of PICC-MARSI injury and possible causes for each case, the specialist team discuss and seek to improve it in the work group, which may also be a reason for the relatively low incidence of MARSI in this study.

Risk factors for PICC-MARSI

The results of the risk factor analysis in this study showed that age >65 years, wet skin, dry skin, oedema, allergy history, dressing type, BMI<18.5 and BMI 25–30 were independent risk factors for PICC-MARSI.

Previous studies have shown that MARSI is related to age, with older adults more prone to MARSI.^{12 15 16} During the ageing process, the skin undergoes a series of changes, including loss of dermal matrix and subcutaneous tissue, thinning of the epidermis, decreased cohesion between the dermis and epidermis, decreased vascular, elastic, and tensile strength, and water loss, which increases the risk of skin injury in older adults.¹⁷

In this study, contact dermatitis was the form of PICC-MARSI with the highest incidence, accounting for 30.3% of all cases. Our risk factor analysis indicated that patients with wet or dry skin had increased risk of contact dermatitis. Obese or sweating-prone patients may have water retention under closed dressings, which can lead to soaking and irritation, making the skin more susceptible to mechanical trauma or dermatitis.¹⁸ More breathable dressings can reduce the risk of PICC-MARSI. In a trial of gauze and polyurethane dressings, IV3000 (Smith & Nephew) had the highest moisture permeability in humid, wet and body temperature simulated environments, but cannot eradicate the increased risk of catheter-related infections, while Tegaderm HP transparent film has advantages in sealing and fixation.¹⁴ Therefore, a personalised and comprehensive selection of dressings is needed, and more versatile dressings are required in clinical practice. In addition, the hospital where this study was conducted is located in a coastal city in southern China, where summer is relatively long and the air humidity is high. Due to sweating, the amount of keratinocytes shed, water-binding and skin wrinkle deepening are higher in summer, further increasing the incidence of irritant dermatitis and soaking.⁴

The incidence of allergic dermatitis was also high in this study, accounting for 21.2% of PICC-MARSI cases. Although some studies have shown that there is a low incidence of allergic dermatitis in MARSI,¹⁹ patients with cancer are more susceptible to allergic dermatitis, due to multiple cycles of chemotherapy or targeted drug therapy. Drugs inhibit the growth of keratinocytes in the epidermis, making the skin barrier fragile,²⁰ and repeated irritation from antisepsis can cause allergic dermatitis in patients. Allergy history was also an independent risk factor for PICC-MARSI, consistent with the results of Zhao et al.¹⁰ As a foreign body placed in the blood vessels, PICC triggers physiological defence mechanisms to produce an immune response, causing skin reactions. In addition, patients with cancer often have hypoalbuminaemia, causing tissue oedema and nutritional disorders.^{21 22} The increased skin tension caused by oedema increases the adhesion force between the skin and the adhesive, which is greater than the adhesive force between the dermis, making patients susceptible to tension blisters and epidermal detachment.²³ Therefore, oedema and BMI<18.5 are independent risk factors for PICC-MARSI in patients with cancer, and improving the nutrition of such patients is crucial.

Limitations

This was a single-centre prospective study. To some extent, the convenience sampling used in this study weakens and limits the representativeness of our results. The same catheter material and maintenance method were used, and all patients with PICC required chemotherapy, which may have led to other risk factors being missed.

Conclusions

Although there are many factors that can induce PICC-MARSI, allergy-prone, sweating-prone physique, and susceptibility related to chemotherapy and targeted drugs are inevitable in patients with cancer. Our findings further emphasise the importance of standardised catheter maintenance operations and the selection of suitable dressings for patients with impaired skin barriers. A cross-sectional descriptive study on CVAD placement site care practices from 34 countries showed that there is a lack of consistency in skin disinfection, dressing selection, dressing replacement frequency and catheter maintenance practices.⁸ Therefore, it is necessary to explore the common and individual problems in CASI in different institutions and develop more suitable catheter maintenance approaches that are targeted and personalised to reduce the incidence of CASI.

Data availability statement

Data are available upon reasonable request. Not applicable.

Ethics statements

Patient consent for publication

Not applicable.

Ethics approval

This study involves human participants and was approved by the Ethics Committee of Huazhong University of Science and Technology Union Shenzhen Hospital (ethics number, KY-2023-092701). Participants gave informed consent to participate in the study before taking part.