C-reactive Protein Levels in Plasma and Chronic Venous Ulcer Exudate of Persons Who Inject Drugs: A Pilot Study

Peer Reviewed

Original Research

C-reactive Protein Levels in Plasma and Chronic Venous Ulcer Exudate of Persons Who Inject Drugs: A Pilot Study

Keywords

C-reactive protein

venous ulcers

Persons Who Inject Drugs

May 2021

ISSN 1044-7946

Index Wounds 2021;33(5):106-112. Epub 2021 March 31. doi:10.25270/wnds/033121.01

Abstract

Background. Persons who inject drugs (PWID) in the groin, legs, and/or feet are at high risk for chronic venous ulcers (CVUs). The plasma C-reactive protein (CRP) level is a marker of systemic inflammation. Objective. This pilot study examined CRP levels in plasma and CVU exudate of PWID. The aims were to (1) compare levels of CRP in plasma and exudate; (2) examine if the CRP level in exudate changed over 4 weeks with wound treatment; and (3) examine the relationship of the exudate CRP level with CVU area, CVU age, number of CVUs, and number of comorbidities. Materials and Methods. Persons who inject drugs seeking wound care were enrolled in this Institutional Review Board approved prospective, longitudinal, descriptive study. A blood sample was collected on the first visit (week 1); the plasma was then separated. Wound exudate was collected on swabs during the first visit (week 1) and 4 weeks later (week 4). All samples were stored at -80° C. Samples were eluted from swabs using mass spectrometry grade water then aliquoted for CRP analysis. Results. The participants of the study included 14 PWID (mean age, 62.14 ± 4.52 years; mean number of comorbidities, 5.71 ± 1.90; and mean number of ulcers 2.07 ± 1.07 that were present for a mean of 7.96 ± 11.91 years without healing). C-reactive protein level in plasma was a mean of 6.47 ± 8.56 mg/L, with lower levels found in wound exudate but highly correlated (rho = .925). Exudate CRP levels decreased from week 1 to week 4, and the 2 were highly correlated (rho = .895). Exudate CRP level week 1 was not significantly related to wound area, wound age, number of ulcers, or number of comorbidities. Conclusions. Plasma and exudate CRP levels were highly correlated. Exudate CRP levels decreased across time. Future large-scale wound healing studies should examine CRP levels over a longer duration and as they correlate to wound healing.

Introduction

In 2011, persons who inject drugs (PWID) comprised 2.6% of the United States population aged 13 years and older, or approximately 6.6 million persons; these numbers have continued to increase.^1,2 Injecting often starts in the upper body; with loss of these peripheral venous sites, PWID generally inject in the veins in the groin, lower extremities, and feet.³ Venous sclerosis and scarring in the lower extremities lead to chronic venous insufficiency with resultant ulcers that may be large and bilateral.^4-6 These ulcers are challenging to treat and difficult to heal. Although inflammatory biomarkers, such as C-reactive protein (CRP), have been examined in venous ulcer healing in the general population,⁷ similar CRP research for PWID could not be found during the present authors’ research. Information about CRP levels in venous ulcers of PWID could lead to insight about wound treatment and prevention in these individuals.

C-reactive protein is an indicator of a systemic inflammatory state.⁸ A pro-inflammatory cytokine synthesized in the liver, CRP increases following interleukin-6 secretion by macrophages and T cells. A reference value for plasma CRP level in persons without comorbidities is any value less than 10 mg/L.⁹ As part of the inflammatory response in wound healing, adipocytes, macrophages, and hepatocytes secrete CRP.¹⁰ Increased hydrostatic pressure in the wound bed of venous ulcers results in the extravasation of neutrophils and macrophages and unrestricted proteolytic activity so that healing is impaired.¹¹ Cushman et al¹² noted CRP level with a subset of other inflammatory markers was associated with an increased risk of peripheral venous disease. Mechanisms that lead to the persistent inflammatory state in nonhealing wounds and the role of various inflammatory cytokines have not been clearly described.¹¹

C-reactive protein levels have been found to be a marker of both the systemic^10,13 and local¹⁰inflammatory states of chronic wounds. Liu et al¹⁰ debrided and treated the wounds of 20 patients with trauma-related, nonhealing chronic wounds with negative pressure wound therapy. Before debridement, the serum CRP level was 22 times higher than the reference range.¹⁰ Serum CRP concentrations decreased significantly with treatment. Trøstrup et al¹⁴ evaluated wound exudate from 25 patients with acute wounds and 8 patients with chronic wounds. The study found significantly elevated CRP levels in chronic wound exudate vs acute wound exudate (P < 0.001).¹⁴ Trengove et al¹⁵ also compared changes in levels of wound exudate cytokines between nonhealing and healing phases from lower extremity wounds of 8 patients. They reported a significant decrease in the CRP level in the healing phase.¹⁵ Beidler et al¹⁶ used multiplex protein assays to assess levels of 22 cytokines in biopsies taken from 30 limbs with untreated chronic venous ulcers (CVUs). Levels of pro-inflammatory cytokines were elevated in ulcer tissue compared with healthy tissue. The levels of cytokines were significantly decreased following 4 weeks of compression therapy¹⁶; unfortunately, the CRP level was not evaluated.

Because plasma and exudate CRP levels in venous ulcers of PWID have not been studied, the purpose of this pilot study was to examine CRP levels in plasma and wound exudate in CVUs of PWID in relation to wound area, wound age, number of CVUs, and number of comorbidities. The aims were to (1) compare levels of CRP in plasma and wound exudate; (2) examine if the CRP level in wound exudate changed over 4 weeks with wound treatment; and (3) examine the relationship of the wound exudate CRP level with CVU area, CVU age, number of CVUs, and number of comorbidities.

Materials and Methods

Design

This pilot study used a prospective, longitudinal, descriptive design that evaluated demographic factors (comorbidities), venous ulcer factors (ulcer area, number of ulcers, and age of ulcers), and plasma and wound exudate CRP levels. The study was funded by Wayne State University, Detroit, Michigan, to collect preliminary data to examine feasibility for future work; as a result, the sample size was small. The study protocol and instruments were reviewed by the university's Institutional Review Board (IRB) and approved (IRB No. 034715M1F). Data were collected from June 2016 to July 2016.

Participants

Participants were recruited from an outpatient clinic that provided chronic wound care primarily to PWID. Individuals were preliminarily screened for eligibility based on age and history of injection drug use. As CVUs can occur at a younger age in PWID,^5,6 eligible participants were to be of 40 years of age and older. Additional inclusion criteria comprised of having a history of injection drug use, presence of a venous ulcer on one or both lower extremities, and ability to verbally respond to the questionnaire items. Participants also needed to agree to have blood drawn on the first visit, have wound exudate collected on swabs on the first visit (week 1) and the fourth visit (week 4), have all wounds assessed, return weekly to the clinic, and follow the wound care treatment protocol. Exclusion criteria included active pregnancy, the inability to respond to the questionnaire due to physical and/or mental illness, lower extremity amputation, and declining the one-time blood draw and/or two-time wound exudate collection. Research assistants involved in the study were trained in proper collection and handling of specimens, administration of questionnaires, and wound assessment and care.

Instruments

Demographic information. The demographic information included injection drug use history as well as general patient information. Sex was listed as male or female. Age was in years. Participants were asked if they had any of 20 comorbidities (eg, high blood pressure, heart failure or congestive heart failure, stroke, diabetes, HIV). They rated their health on a numeric scale ranging from a score of 1 (ill) to 10 (healthy).

Injection drug use history included total years injecting as well as years injecting in the groin, legs, and/or feet. Participants were asked to identify either the age or year they first injected street drugs as well as the age or year of last injection; the total years of injection use were determined based on the difference between the most recent injection and initial injection.¹⁷ Subsequently, participants were asked if there were years in which they had not injected during the overall timeframe of injected substance use. Years not injecting were removed from the total years of injecting.¹⁷ Additionally, participants were asked how many of the total injecting years included injecting in the groin, legs, and/or feet.

Venous ulcer age. Participants were asked for the year the ulcer developed. The age of the venous ulcer was calculated by subtracting that year from the current year. If the ulcer developed during the year of the study, the age of the ulcer was recorded as a decimal number per the months of occurrence.

Venous ulcer area. The area of the venous ulcer was calculated by measuring the ulcer with a disposable ruler using the greatest length and width (in cm). The total ulcer area was calculated by multiplying the length by width measurements. If more than one ulcer was present, the total wound area was represented by the sum of all the ulcer areas. Number of ulcers was determined by counting each participant’s venous ulcers.

Blood and wound exudate. After the wound dressing was removed and the wound was flushed with tap water, wound exudate was obtained by placing a cotton-tipped swab (Copan Italia S.p.A.) in the wound and securing the swab in place with a lightly wrapped Kerlix (Covidien). With legs hanging down, after 10 minutes, the swab was removed, the swab was removed, capped, immediately placed on ice, and stored frozen at -80° C. If more than one venous ulcer was present, wound exudate was obtained from each ulcer. Blood was drawn from the antecubital space or another location if that site was not possible. Blood (3 ml in 2 EDTA tubes) was obtained on the first visit (week 1). Because of difficulty in obtaining blood from PWID (due to venous trauma and scarring from injected drugs) and the amount of funding available, a blood sample was obtained once. The plasma was separated by centrifuging blood collection tubes at 10000 x g for 5 minutes; it was then stored frozen at -80° C. Samples were eluted from swabs using mass spectrometry grade water and then aliquoted for cytokine analysis. Enzyme-linked immunosorbent assay analysis was used for CRP level in both plasma and wound exudate samples. A normal value for CRP level is less than 10 mg/L,⁹ with circulating levels greater than 3 mg/L used to identify elevated risk of cardiovascular disease.¹⁸ A normal range for CRP level in wound exudate has not been established. The AlphaLisa (PerkinElmer) assay was used to quantify CRP levels. The assay has a lower detection limit of 5.62 picograms per mL, and the percent coefficient of variation for replicate measures was 6.33%.

Procedure

The study was described to eligible and interested participants; signed consent was obtained. After obtaining consent, a research assistant read the questionnaires to participants to facilitate completion of the questionnaire packet. The participants’ lower extremity dressings were removed. The wounds were washed with tap water and then measured. Measurements were taken at the greatest length vertical to the head and at the greatest horizontal width in centimeters, and wound exudate was obtained twice during the 4-week period (at week 1 and week 4). Routine wound care using compression therapy was done at week 1 and throughout the study. No participant received advanced therapy because it was not available to patients of this clinic.

The research assistant accompanied each participant to the laboratory for blood collection at week 1. Participants were compensated $30 for the first visit (week 1) and $20 for the last visit (week 4).

Statistical analysis

Responses to the questionnaire were entered in SPSS Statistics Version 26 (IBM). Descriptive statistics were used to describe the sample in terms of sex, age, comorbidities, ulcers, and injection drug use history. Correlations were examined with Spearman rank correlation (rho) for plasma versus wound exudate CRP levels and with demographic and wound variables. The non-parametric Spearman rho was used to evaluate relationships between variables for rank order, considering the small number of participants.

Results

Participants

The 14 participants included 9 (64%) males; age ranged from 56 years to 70 years (mean, 62.14 years; SD, 4.52). Participants had a mean of 5.71 comorbidities (SD, 1.89), and the mean health average rating was 5.86 (SD, 1.35). Duration of injecting drugs in the groin, feet, and/or legs ranged from 1 year to 48 years (mean, 19.64 years; SD, 15.10). Seven participants had venous ulcers on both legs, 5 had ulcers located on the left leg, and 2 had ulcers located on the right leg. The mean duration of having the largest current leg ulcer was 7.96 years (SD, 11.91). The mean total wound area at week 1 was 59.85 cm² (SD, 47.74 cm²). Data for individual participants are in Table 1.

C-reactive protein levels in plasma and wound exudate

The week 1 mean CRP level in plasma was 6.47 mg/L, while the mean CRP level in wound exudate was 0.42 mg/L (0 – 1.77 mg/L; P = .015) (Table 1). A strong correlation was found between week 1 CRP level in plasma and CRP level in wound exudate (rho = .925; P < .001) (Figure 1).

Exudate C-reactive protein level over time

Values of CRP levels in wound exudate for week 1 and week 4 were found to be highly correlated (rho = .895; P < .001) (Figure 2). Mean exudate CRP levels decreased significantly over the 4-week period from 0.42 mg/L to 0.29 mg/L (one-sided related-sample Wilcoxon signed rank test P = .04) (Figure 3).

Week 1 wound exudate

The correlations among wound exudate week 1 CRP level and wound area, wound age, number of ulcers, and number of co-morbidities are depicted in Table 2. None were significant.

Discussion

This pilot study examined the CRP levels in plasma and the wound exudate amount in CVUs for PWID in relation to the wound area and age and the number of CVUs and comorbidities. Venous ulcer parameters of PWID are infrequently studied. The first aim of the present study was to compare levels of CRP in plasma and wound exudate.

To date, underlying mechanisms of wound healing related to CRP levels have not been completely determined.¹⁹ According to Lindley et al,²⁰ while plasma CRP levels can give an indication of systemic inflammation, CRP levels in wound exudate can provide an indication of local inflammation. Liu et al¹⁰ proposed that wound healing can be mediated by resolving the inflammatory state. As plasma CRP levels give an indication of inflammation somewhere in the body, CRP levels in wound exudate provide another potential indicator of healing with definitive interpretation of the changing wound fluid CRP levels depending upon further study. The present authors found that plasma CRP levels correlated well with wound exudate CRP levels in this group. This is consistent with results from Trengove et al,¹⁵ who found that levels of CRP in wound exudate (8.5 mg/L) were similar to levels of CRP in serum (6.4 mg/L) (P > .05). Fivenson et al²¹ found higher values of tumor necrosis factor alpha and interleukin-6 in plasma than in wound exudate; however, CRP values were not assessed. Gohel et al²² found that cytokine concentrations in wound exudate and serum correlated poorly, but CRP was not considered.

The present authors found CRP levels in wound exudate correlated well with CRP levels in plasma; thus, wound exudate can potentially be collected and analyzed in lieu of blood draws, which can cause the patient discomfort or pain and be difficult to obtain because of venous trauma and scarring from injected drugs. Blood draws for PWID are especially challenging due to the scarring and sclerosing of their veins.¹⁷ There are inconsistencies in the literature for collecting wound exudate,²³ but the present authors found collection on swabs provided sufficient wound exudate for analysis of CRP levels. This finding may be useful for other researchers who want varied substances that are present both in wound exudate and plasma. However, more research is needed in regard to this method.

The second aim of the present study was to examine if CRP levels in wound exudate changed with wound treatment over the course of 4 weeks. C-reactive protein levels in wound exudate decreased significantly over time, while wound area did not. A decrease in wound area is often related to healing.⁴ O’Donnell et al²⁴ reported that surface area measurements are recommended to determine wound healing. However, Stacey et al²⁵ noted there is not a consistent way to measure healing week by week. Liu et al¹⁰ found CRP concentrations in wound exudate decreased significantly with negative pressure wound therapy. As previously noted, Beidler et al¹⁶ found that levels of cytokines were significantly decreased following 4 weeks of compression therapy, but CRP was not mentioned. Trengove et al¹⁵ found that levels of CRP in wound exudate decreased significantly between nonhealing and healing phases of wounds.

The authors’ third aim was to examine the relationships among wound exudate CRP level with CVU area, CVU age, number of CVUs, and number of comorbidities. Significant results were not found when correlating CRP levels in wound exudate with wound area, wound age, number of ulcers, or number of comorbidities. The large values of the demographic factors (including years of injecting, numbers of wounds, and numbers of comorbidities) in Table 1 indicate the challenges in providing wound care to PWID. The age of the participant significantly correlated with wound age and wound area in this study, indicating that persons older than 60 years are likely to have more and longer-lasting CVUs. As PWIDs age, there is greater potential for comorbidities and a longer history of wounds than in younger individuals.

Although the authors did not examine psychosocial factors, psychosocial factors have been recognized as impacting wound healing in PWID (eg, self-wound care, distrust of the health care system, not returning for appointments).^4,26-29 Research is needed to determine the best way to assess wound healing in PWID. Recent media coverage of the opioid crisis enhances the need for continued exploration of wound assessment and healing for PWID.

Limitations

This study has several limitations. Due to funding limitations, the number of participants was small (N = 14). Thus, the small number of participants limits generalizability and results need to be replicated in a larger sample. Another limitation is that the participants had been receiving ongoing care at week 1 and were not new patients to the clinic. To meet the time restrictions of the study, it was necessary to recruit participants from the current patient population of the clinic. Results might have been more pronounced if patients had been recruited at initiation of treatment.

Of note, all participants were placed in compression wraps. The clinic did not have access to advanced wound treatments (advanced therapies were not provided because they were not available due to contract issues), thus the effect of advanced treatments on exudate and plasma CRP levels as well as healing could not be determined in the present study. While all participants had a history of injection drug use, the authors did not have CRP levels from persons who did not inject drugs, thereby making a comparison of the two impossible to obtain in the current study. Participants were provided a non-excessive monetary compensation, which may have contributed to keeping the participants engaged in the study.

Conclusions

C-reactive protein levels in plasma and wound exudate were highly correlated in PWID with venous ulcers. The CRP wound exudate values significantly decreased from week 1 to week 4. Results from this pilot study indicate that determining CRP levels in wound exudate may be a useful alternative to obtaining blood for plasma CRP levels in wound studies, particularly given the venous damage associated with PWID. Larger studies are needed to confirm these results in PWID and persons who do not inject drugs.

Acknowledgements

Authors: Julia Paul, PhD, RN, ACNS-BC, CCRN, CWS^1,2; Thomas N. Templin, PhD³; Nicholas J. Carruthers, PhD⁴; Paul R. Burghardt, PhD⁵; Ciara Ivanics, MS⁶; Paul M. Stemmer, PhD⁴; and Barbara Pieper, PhD, RN, CWOCN, ACNS-BC, FAAN³

Affiliations: ¹School of Nursing, Oakland University, Rochester, MI; ²William Beaumont Hospital, Royal Oak, MI; ³College of Nursing, Wayne State University, MI; ⁴Institute of Environmental Health Sciences, Wayne State University; ⁵Department of Nutrition and Food Sciences, Wayne State University; ⁶School of Medicine, Wayne State University

Correspondence: Paul M. Stemmer, PhD, Institute of Environmental Health Sciences, Wayne State University, 259 Mack Avenue, Detroit, MI 48201; ak5245@wayne.edu; Barbara Pieper, PhD, RN, CWOCN, ACNS-BC, FAAN, College of Nursing, Wayne State University, 5557 Cass Avenue, Detroit, MI 48202; ab5813@wayne.edu

Disclosure: This study was funded by a Wayne State University BOOST Award. The authors disclose no additional conflicts of interest.

References

1. Jones CM, Logan J, Gladden M, Bohm MK. Vital signs: demographic and substance use trends among heroin users—United States, 2002–2013. MMWR. 2015;64(26):719–725.

2. Lansky A, Finlayson T, Johnson C, et al. Estimating the number of persons who inject drugs in the United States by meta-analysis to calculate national rates of HIV and hepatitis C virus infections. PLoS One. 2014;9(5):e97596. doi:10.1371/journal.pone.0097596

3. Ciccarone D, Harris M. Fire in the vein: heroin acidity and its proximal effect on users’ health. Int J Drug Policy. 2015;26(11):1103–1110. doi:10.1016/j.drugpo.2015.04.009

4. Pieper B. A retrospective analysis of venous ulcer healing in current and former users of injected drugs. J Wound Ostomy Continence Nurs. 1996;23(6):291–296. doi:10.1016/s1071-5754(96)90048-0

5. Pieper B, Kirsner RS, Templin TN, Birk TJ. Injection drug use: an understudied cause of venous disease. Arch Dermatol. 2007;143(10):1305–1309. doi:10.1001/archderm.143.10.1305

6. Pieper B, Templin TN, Kirsner RS, Birk TJ. Impact of injection drug use on distribution and severity of chronic venous disorders. Wound Repair Regen. 2009;17(4):485–491. doi:10.1111/j.1524-475X.2009.00513.x

7. Ligi D, Mosti G, Croce L, Raffetto JD, Mannello F. Chronic venous disease – part I: inflammatory biomarkers in wound healing. Biochim Biophys Acta. 2016;1862(10):1964–1974. doi:10.1016/j.bbadis.2016.07.018

8. Iizaka S, Sanada H, Minematsu T, et al. Do nutritional markers in wound fluid reflect pressure ulcer status? Wound Rep Regen. 2010;18(1):31–37. doi:10.1111/j.1524-475X.2009.00564.x

9. Mayo Foundation for Medical Education and Research. C-reactive protein test. Mayo Clinic. Accessed February 28, 2021. https://www.mayoclinic.org/tests-procedures/c-reactive-protein-test/about/pac-20385228?p=1

10. Liu T, Yang F, Li Z, Yi C, Bai X. A prospective pilot study to evaluate wound outcomes and levels of serum C-reactive protein and interleukin-6 in the wound fluid of patients with trauma-related chronic wounds. Ostomy Wound Manage. 2014;60(6):30–37.

11. Willenborg S, Knipper J, Ranjan R, Krieg T, Eming SA. Chronic wounds and inflammation. In: Sen CK, ed. Advances in Wound Care. Volume 1. Mary Ann Liebert, Inc; 2010:259–265.

12. Cushman M, Callas PW, Allison MA, Criqui MH. Inflammation and peripheral venous disease. The San Diego population study. Thromb Haemost. 2014;112(3):566–572. doi:10.1160/TH13-10-0860

13. Lassig AAD, Bechtold JE, Lindgren BR, et al. Tobacco exposure and wound healing in head and neck surgical wounds. Laryngoscope. 2018;128(3):618–625. doi:10.1002/lary.26813

14. Trøstrup H, Lundquist R, Christensen LH, et al. S100A8/A9 deficiency in nonhealing venous leg ulcers uncovered by multiplexed antibody microarray profiling. Br J Dermatol. 2011;165(2):292–301. doi:10.1111/j.1365-2133.2011.10384.x

15. Trengove NJ, Langton SR, Stacey MC. Biochemical analysis of wound fluid from nonhealing and healing chronic leg ulcers. Wound Repair Regen. 1996;4(2):234–239. doi:10.1046/j.1524-475X.1996.40211.x

16. Beidler SK, Doulliet CD, Berndt DF, Keagy BA, Rich PB, Marston WA. Inflammatory cytokine levels in chronic venous insufficiency ulcer tissue before and after compression therapy. J Vasc Surg. 2009;49(4):1013–1020. doi:10.1016/j.jvs.2008.11.049

17. Pieper B, Templin TN, Birk TJ, Kirsner RS. Reliability and clinical validity of a technique to assess lifetime illicit drug use. Ostomy Wound Manage. 2008;54(2):16–34.

18. Ledue TB, Rifai N. Preanalytic and analytic sources of variations in C-reactive protein measurement: implications for cardiovascular disease risk assessment. Clin Chem. 2003;49(8):1258–1271. doi:10.1373/49.8.1258

19. Kalkhof S, Förster Y, Schmidt J, et al. Proteomics and metabolomics for in situ wound healing. Biomed Res Int. 2014;2014:934848. doi:10.1155/2014/934848

20. Lindley LE, Stojanovic O, Pastar I, Tomic-Camic M. Biology and biomarkers of wound healing. Plast Reconstr Surg. 2016;138(3 Suppl):18S–28S. doi:10.1097/PRS.0000000000002682

21. Fivenson DP, Faria DT, Nickoloff BJ, et al. Chemokine and inflammatory cytokine changes during chronic wound healing. Wound Repair Regen. 1997;5(4):310–322. doi:10.1046/j.1524-475X.1997.50405.x

22. Gohel MS, Windhaber RAJ, Tarlton JF, Whyman MR, Poskitt KR. The relationship between cytokine concentrations and wound healing in chronic venous ulceration. J Vasc Surg. 2008;48(5):1272–1277. doi:10.1016/j.jvs.2008.06.042

23. Mannello F, Ligi D, Canale M, Raffetto JD. Omics profiles in chronic venous ulcer fluid: innovative applications for translational medicine. Expert Rev Mol Diagn. 2014;14(6):737–762. doi:10.1586/14737159.2014.927312

24. O’Donnell TF, Passman MA, Marston WA, et al. Management of venous leg ulcers: clinical practice guidelines of the Society for Vascular Surgery® and the American venous forum. J Vasc Surg. 2014;60(2 Suppl):3S–59S. doi:10.1016/j.jvs.2014.04.049

25. Stacey MC, Phillips SA, Farrokhyar F, Swaine JM. Evaluation of wound fluid biomarkers to determine healing in adults with venous leg ulcers: a prospective study. Wound Repair Regen. 2019:27(9):509–518. doi:10.1111/wrr.12723

26. Coull AF, Atherton I, Taylor A, Watterson AE. Prevalence of skin problems and leg ulceration in a sample of young injecting drug users. Harm Reduct J. 2014;11:22. doi:10.1186/1477-7517-11-22

27. Pirozzi K, Van JC, Pontious J, Meyr AJ. Demographic description of the presentation and treatment of lower extremity skin and soft tissue infections secondary to skin popping in intravenous drug abusers. J Foot Ankle Surg. 2014;53(2):156–159. doi:10.1053/j.jfas.2013.12.017

28. Roose RJ, Hayashi AS, Cunningham CO. Self-management of injection-related wounds among injecting drug users. J Addict Dis. 2009;28(1):74–80. doi:10.1080/10550880802545200

29. Shiman MI, Pieper B, Templin TN, Birk TJ, Patel AR, Kirsner RS. Venous ulcers: a reappraisal analyzing the effects of neuropathy, muscle involvement, and range of motion upon gait and calf muscle function. Wound Repair Regen. 2009;17(2):147–152. doi:10.1111/j.1524-475X.2009.00468.x