1 INTRODUCTION

Chronic leg ulcers (CLUs) are hard-to-heal wounds localised on the lower leg or foot, persisting for more than 6 weeks.1 CLUs are common, and affect approximately 0.1% to 2% of the adult population in Western countries.2 Ulcer related pain is a significant problem, and all CLUs are potentially painful.3-5 Pain is defined as ‘an unpleasant sensory and emotional experience associated with, or resembling that associated with, actual or potential tissue damage’.6 Ulcer related pain can be classified as background, incident, procedural, and operative.7 Background pain is caused by the underlying pathology of the leg ulceration and the wound itself, incidence pain stems from various daily activities, while procedural and operative pain is caused by wound treatment and debridement.8 From a clinical perspective, it is probably difficult for persons with CLUs to separate the experience of background pain from incidence pain.9 Hence, in this study, background pain is defined as ulcer related pain experienced in everyday life (other than procedural/operative pain).9 Ulcer related pain is often multi-causal in nature, and can be further classified as nociceptive, neuropathic, inflammatory, as well as acute and/or chronic.10-13 In addition, persons with CLUs are often old and may have multiple painful comorbidities.14, 15 Consequently, persons with CLUs can have very complex pain conditions with the ulcer related background pain as an additional problem on top of living with a CLU and other painful comorbidities.

In the beginning of this century, several consensus statements based on expert opinion drew attention to the extent – and treatment of ulcer related pain.5, 7, 16 And the well-known concept of holistic wound bed preparation paradigm emphasises patient comfort and relieving pain in wound care.17 Effective pain management in wound care depends on detailed and accurate assessment and documentation of the pain experience, and a mix of psychosocial approaches together with local and systemic pain management should be implemented to control it.17, 18 Most research studies on pain management in wound care have focused on the effects of different low-adherent or non-adherent dressings and topical analgesics and local anaesthetic agents mainly aimed to reduce dressing related pain.19, 20 Larger studies assessing more extensive and holistic management are not available. However, despite the publication of clinical guidelines, consensus documents and research, ulcer related background pain is still not effectively assessed nor addressed in leg ulcer care,21, 22 and background pain continues to be a frequent and substantial problem.23

Most research studies on CLUs report on diagnosis, prevention, treatment, and healing rates.24, 25 Few studies report on ulcer related pain, and this research often concentrates on acute procedure-related pain and pain at dressing change. Knowledge is therefore scarce on the characteristics of ulcer related background pain. Intensity is the most commonly reported pain characteristic, and mean background pain intensity in CLUs is estimated at 3 to 4 (0-10 numeric rating scale).9, 23, 26 A limited number of studies describe other characteristics of ulcer related pain than intensity, such as pain descriptors, interference, and temporal pattern. Throbbing, tender, and burning were the most common descriptors of background pain.26-28 Sensory descriptors were typically used more often than affective descriptors.26, 29 Ulcer related background pain can interfere with sleep,30, 31 and one study found moderate pain interference in a sample with venous leg ulcers.32 Finally, background pain was worse during standing than at rest, and intermittent pain was more frequent than continuous pain.33 Qualitative studies also support the use of sensory descriptors, varying intensity, and pain interference.34 However, no previous studies provide a methodologically sound and comprehensive description of several characteristics of ulcer related background pain in a large sample of persons with CLUs.

CLUs are a substantial problem, and ulcer related background pain is a frequent and bothersome symptom disrupting quality of life. Knowledge about characteristics of ulcer related background pain is scarce and little is known about the pain management persons with CLUs receive. A thorough evaluation of the pain symptomatology in these persons is necessary for a better understanding of the pain experience and addressing the pervasive impact on function. Hence, the aim of the current study was to explore and describe characteristics of ulcer related background pain and pain management as reported by persons with CLUs.

2 METHODS 2.1 Design and participants

The current study is a descriptive analysis of cross-sectional data. The data collection was conducted from May 2017 to the end of December 2018. Consecutive sampling method was used to recruit participants attending two outpatient clinics for wound care in South-Eastern Norway. Inclusion criteria were: (a) presence of an open wound located below the knee; (b) wound duration of 6 weeks or longer; (c) age 18 years or older, (d) ability to understand and read Norwegian; and (e) no cognitive impairment causing comprehension difficulties. Exclusion criteria included the following ulcer diagnosis: burn-ulcer, cancer-ulcer, radiation-ulcer, pressure ulcers caused by immobility, and immunological ulcers.

The present study sample was selected from a larger sample of persons with CLUs (reference deleted for blinded review). To explore pain characteristics, we analysed data from 121 persons who reported presence of ulcer related background pain in the screening interviews, and who returned the self-report questionnaires.

2.2 Data collection

Data were gathered with an initial screening interview, and a clinical examination at the wound outpatient clinic, as well as a self-report questionnaire filled in at home within 24 hours after the hospital visit.

2.2.1 Screening interview

The initial brief interview was conducted by LL before the scheduled appointment at the wound care clinic. Participants were asked for the presence and type of ulcer related pain, and a structured questionnaire was used to gather data on wound duration and reoccurrence, as well as comorbidities.

2.2.2 Clinical examination

During the scheduled appointment at the wound clinic, an examination was conducted by LL. Data were collected on ulcer characteristics, sensibility of the foot, location and temporal pattern of ulcer related pain, and pain management. Both participants self-report and patient records of pain management were incomplete in some cases. To ensure consistency of the clinical test, all tests were performed by one researcher LL.

Wound characteristics: The medical doctor assessed the ulcer and stated the ulcer diagnosis (ie, venous, diabetic, traumatic, unspecified). In addition, the researcher LL asked the participants about the presumed causal factor (eg, trauma, pressure), and identified and marked the localization of the ulcer on a body map of the lower leg. Wound size was estimated by multiplying the width and height (millimetres at the widest and highest) of the wound, calculating the area using the formula of an ellipse.

Sensibility of the lower leg: Tactile static mechanical sensation was assessed using Semmes-Weinstein monofilament (5.07/10 g) examination (SWME).35 Two pricks with the monofilament were applied on each of four plantar sites on the forefoot (great toe, and base of first, third, and fifth metatarsals). Inability to detect one prick is used as a diagnostic threshold when screening for diabetic peripheral neuropathy.36 Vibration sense was assessed using a 128-Hz tuning fork. The vibrating tuning fork was first placed on bony prominence where neuropathy is unlikely (eg, hand). Once the participant was familiar with the vibration, the vibrating fork was applied to the bony prominence situated at the dorsum of the first toe just proximal to the nail bed.37, 38 The respondents indicated whether they could feel the vibration with their eyes closed. Reduced or lost sense of vibration stimuli may indicate neuropathy.39

Location of pain: Participants were asked to identify the pain location, such as directly in the wound bed; in the wound edges; in the area surrounding the wound; in the entire foot; in the entire leg, or a self-determined location. If more than one location was selected, ‘multiple locations’ were used to describe the location of the pain.

Temporal pattern of pain: To assess how pain fluctuated during the day, a question from the pain quality assessment scale (PQAS)40 was used. PQAS is a valid tool for assessing various types of nociceptive and neuropathic pain,40, 41 and is linguistically validated and culturally adapted in Norwegian.42 The participants were asked whether their pain was intermittent (feel pain sometimes, pain free at other times), variable (pain all the time, but also moments of more severe/different pain), or stable (constant pain that does not change).

Pain management: Participants were asked about their use of analgesics, and compliance to prescribed analgesics. Supplemental/confirmatory information was gathered from the patients' records when available. It was differentiated whether the pain management was aimed at ulcer related or other types of pain. Finally, participants were asked to describe other non-medical interventions used to relieve ulcer related pain.

2.2.3 Questionnaires

All participants received a battery of self-report questionnaires on the day of recruitment and scheduled appointment at the wound clinic. The questionnaires were completed the following day and returned by mail in a pre-paid envelope. The questionnaires contained questions regarding demography and pain characteristics. Participants were specifically asked to keep in mind their ulcer related pain when answering questions regarding pain.

Demography: Information was obtained on age, gender, work situation, education, living arrangements, and ethnicity.

Short-form McGill pain questionnaire (SF-MPQ) was used to assess qualities of present ulcer related background pain. The SF-MPQ contains 15 pain descriptors (11 sensory, four affective) with a four-point intensity scale for each word (ie, none, mild, moderate, severe). In addition, the SF-MPQ contains a present pain intensity index and a 0 to 100 mm visual analogue scale (VAS) to assess present pain intensity.43 The Norwegian version of SF-MPQ has demonstrated reliability and validity in a Norwegian sample with musculoskeletal pain.44, 45 The English version of SF-MPQ has been used to assess ulcer related pain.46 The SF-MPQ was linguistically adjusted to specifically assess ulcer related pain by adding ‘ulcer related’ to the word ‘pain’ throughout the questionnaire. The Cronbach's alpha for all the descriptors were 0.87, and the Cronbach's alpha for the sensory and affective descriptors were 0.84 and 0.68, respectively.

The brief pain inventory (BPI) was used to provide information about the location, intensity, treatment, and interference of ulcer related pain on function in the last 24 hours. BPI consists of four 0 (no pain) to 10 (worst pain imaginable) numeric rating scales (NRS) on the intensity of pain in general, at its worst, at its least, and right now. A percentage scale (0-100) quantifies pain relief from current therapies, and a bodymap is provided to mark the pain location. Furthermore, seven questions address whether pain interferes with physical and psychosocial function.47 The activity cluster of interference items (WAW) includes general activity, walking ability, work, and sleep, while the affect cluster of interference items (REM) includes mood, enjoyment of life, and relations with others. The translated Norwegian version of the BPI was validated for use in patients with cancer and osteoarthritis, demonstrating good validity and reliability.48-51 The English version of the BPI was used to assess ulcer related pain.52 The BPI was linguistically adjusted to specifically assess ulcer related pain by adding ‘ulcer related’ to the word ‘pain’ throughout the questionnaire. Cronbach's alpha for both the activity cluster and the affect cluster of interference items were 0.89, and the Cronbach's alpha for all interference items was 0.93 in this study.

Avoidance of physical activities: The participants were asked if they avoided certain activities because of ulcer related pain using a yes/no format. If yes, the participants were asked to describe the activities in own words.

2.3 Research ethics

Approval for the study was granted by the Norwegian Regional Ethical Comity for Medical and Health Research Ethics, region South-East (REK number 2016/1236). In addition, the Norwegian Center for Research Data (NSD), as well as the research manager and the head of department at the local hospitals, approved the study. Participants were informed both orally and written about the aims and procedures of this study, and written consent was obtained from all participants. They were informed that the information they provided would be deidentified, and they could withdraw from the study at any time before the publication of the study.

2.4 Data analysis

SPSS version 26 was used to perform descriptive statistical analyses. Categorical data are presented as frequencies with proportions, and continuous data are presented as means with standard deviation (SD) or medians with range. Chronbach's alpha was calculated for assessing the reliability of the BPI interference sub-scales and the NSF-MPQ total, affective, and sensory descriptors.

BPI and MPQ-SF were both scored according to the developers' recommendations.43, 53 In the analysis, the BPI item number three (worst pain intensity) was transformed to an ordinal variable where 0 to 3.99 correspond to no/mild pain, ≥4 to 6.99 correspond to moderate pain, and ≥7 to 10 correspond to severe pain. These cut-off points are in line with previous studies.23, 54, 55

Responses to open-ended questions were quantified into categories. Categories for avoidance of physical activity were walking, household chores, social activities, bending/crouching, and sitting with legs down. Categories for non-medical interventions were elevating the feet, activity, reduced activity, lowering feet, and other (ie, massage, ulcer care, aids to protect the ulcer).

There were few missing values on single items overall in the dataset. However, one single item, the SF-MPQ VAS present pain intensity item, had 27.3% missing values. Missing items on the 15 pain descriptors of SF-MPQ were replaced by 0. By experience, participants skip marking the pain descriptors that are not relevant to them, causing an inaccurate and high number of missing items.

3 RESULTS 3.1 Sample characteristics

The mean age of participants was 74.4 years (SD 12.75), and 53.7% were female. A total of 39.7% were living alone, and 87.5% were not working (ie, on sick leave, retired) (Table 1).

TABLE 1. Demographic data (N = 121) N (%) Age <49 5 (4.1) 50-59 11 (9.1) 60-69 17 (14.0) 70-79 44 (36.4) 80-89 34 (28.1) >90 10 (8.3) Gender Male 56 (46.3) Female 65 (53.7) Marital status Single 19 (15.7) Married/cohabitant 72 (59.5) Widowed 30 (24.8) Educational level Primary school 48 (39.7) Secondary school 37 (30.6) University <4 years 24 (19.8) University >4 years 9 (7.4) Work situation Working 13 (10.8) Retired/sick leave 106 (87.5) Living arrangements Alone 48 (39.7) With spouse/other family 73 (60.3) Ethnicity Norwegian 118 (97.5) Other 2 (1.7) Note: N, number of participants. 3.2 Clinical characteristics

The participants had at least one active ulcer located at the leg, ankle, or foot. Unspecified ulcer in lower extremities (ICD-10 diagnosis code L97) was the most frequent diagnosis (30.6%). Most ulcers (48%) were triggered by a trauma, but most ulcers were further diagnosed based on physiological factors such as underlying diseases and vascular conditions. The mean wound size was 4.2 cm2 (SD 8.8), and mean wound duration was of 28.5 weeks (SD 47.8) (Table 2).

TABLE 2. Clinical characteristics (N = 121) Median (IQR) N (%) Wound diagnosis (as specified by doctor) Unspecified (L97) 37 (30.6) Venous 25 (20.7) Diabetic foot ulcer 20 (16.5) Traumatic 17 (14.0) Arterial 9 (7.4) Other 13 (10.8) Wound size (cm2) 1.2 (0.46-4.32) Wound duration (weeks) 15 (8-26) Reoccurring wound Yes 55 (45.5) No 57 (47.1) Number of wounds 1 62 (51.2) 2-5 38 (31.4) >5 12 (9.9) Presumed primary causal factor Trauma 123 (48) Pressure/friction 50 (19.8) Venous insufficiency 16 (6.3) Arterial insufficiency 5 (2) Unknown/other 58 (23) Wound location Leg 44 (36.4) Foot 36 (29.8) Ankle 23 (19.0) Multiple locations 18 (14.9) Number of comorbidities 2 (1.5-3) No of comorbidities 6 (5.0) One comorbidity 24 (19.8) Two comorbidities 35 (28.9) Three comorbidities 30 (24.8) Four or more comorbidities 26 (21.5) Comorbidities Coronary disease 90 (74.4) Other painful conditionsa 63 (52.1) Diabetes 37 (30.6) Malnutrition 25 (20.7) Arthritis 22 (18.2) Renal disease 20 (16.5) Cancer (previous or present) 20 (16.5) Stroke 12 (9.9) Asthma/COPD 8 (6.6) SWME 0-3/8 42 (34.7) 4-7/8 29 (23.9) 8/8 42 (34.7) Tuning fork test Positive 62 (51.2) Negative 49 (40.5) Abbreviations: IQR, interquartile range; SD, standard deviation; SWME, Semmes-Weinstein monofilament examination.

For 46%, the present ulcer was not the first CLU they suffered from. A total of 94% reported at least one comorbidity. The most frequent comorbidities were coronary diseases (74%) including hypertension and arteriosclerosis, other painful conditions (52%) such as musculoskeletal pain, and diabetes (31%). Loss of tactile and vibration sensation was common: 59% obtained >8 on the SWME, and 51% did not detect vibration from the tuning fork (Table 2).

3.3 Types of pain

In this sample, all participants reported having ulcer related background pain. One in five reported having only ulcer related background pain, while four in five reported both background pain and pain at dressing change (procedural) (Figure 1).

Background pain and pain at dressing change

3.4 Characteristics of ulcer related background pain

Ulcer pain intensity: The mean average pain intensity was 4.5 (SD 2.56) (CI 95% 4.0-5.0), and the mean worst pain intensity was 4.9 (SD 2.88) (CI 95% 4.4-5.5) (0-10 NRS). Mean present pain intensity was 38.65 mm (SD 27.23) (0-100 VAS) (Table 3). Divided into categories, 37% reported no or mild pain (NRS 0-3), 28% moderate pain (NRS 4-6), and 35% severe pain (NRS 7-10) (Figure 2).

TABLE 3. Ulcer pain characteristics (N = 121) N (%) Mean SD BPI pain intensity (0-10 NRS, past 24 hours) Worst 4.9 2.88 Average 4.5 3.56 Now 3.0 3.64 Least 2.1 2.04 BPI pain interference (0-10 NRS, past 24 hours) General activity 4.3 3.16 Sleep 4.1 3.18 Walking ability 4.0 3.11 Normal work 3.9 3.14 Enjoyment of life 3.6 3.09 Mood 3.3 2.96 Relations with others 2.4 2.69 Activity pain interference (WAW) 4.1 2.8 Affective pain interference (REM) 3.1 2.7 Total interference score 3.6 2.6 SF-MPQ Descriptors SF-MPQ sensory 6.90 6.66 SF-MPQ affective 1.36 2.19 SF-MPQ total 8.24 8.35 VAS present pain intensity (0-100) 38.65 27.23 Present pain intensity index No pain 10 (8.3) Mild pain 22 (18.2) Discomforting 28 (23.1) Distressing 45 (37.2) Horrible 7 (5.8) Excruciating 2 (1.7) PQAS pain pattern Intermittent pain 86 (71.1) Variable pain 31 (25.6) Stable pain 0 (0) Localization of ulcer related pain Multiple locations 69 (57.0) Only the wound itself (wound bed or edges) 27 (22.3) The wound bed 90 (74.4) The wound edges 15 (12.4) The area surrounding the wound 67 (55.4) The entire foot 7 (6.0)