Perceived stress and cortisol levels predict speed of wound healing in healthy male adults

https://doi.org/10.1016/S0306-4530(03)00144-6Get rights and content

Abstract

The main purpose of the present study was to investigate the association between perceived stress and impaired cutaneous wound healing in humans using a novel wound assessment technique, and taking into account putative mediating factors such as cortisol levels, health behaviours, and personality factors.

The study made use of a prospective, within-subjects design in which 24 male non-smokers participated. Every subject received a standard 4mm-punch biopsy, and the healing progress was monitored via high-resolution ultrasound scanning. Participants completed questionnaires on perceived stress, health behaviours, and personality factors, and sampled saliva for cortisol assessment after awakening at 2 weeks prior, directly after, and 2 weeks after the biopsy.

The overall results showed a significant negative correlation between speed of wound healing, and both Perceived Stress scale (PSS) scores (r=−.59; p<.01), and General Health Questionnaire (GHQ) scores (r=−.59; p<.01) at the time of the biopsy. The area under the morning cortisol response curve was negatively correlated with speed of wound healing (r=−.55; p<.05), indicating a clear elevation in the morning cortisol slope of those whose wounds were slowest to heal. A median split of the complete sample yielded that the ‘slow healing’ group showed higher stress levels (PSS t=3.93, p<.01, GHQ t=2.50, p<.05), lower trait optimism (t=3.25, p<.05), and higher cortisol levels to awakening (F=5.60, p<.05) compared with the ‘fast healing’ group. None of the health behaviours investigated (i.e. alcohol consumption, exercise, healthy eating, and sleep) were correlated with healing speed at any time point.

Our data hint at a considerable influence of stress on wound healing, and suggests that elevated cortisol levels, rather than altered health behaviours, play a role in this effect.

Introduction

Cutaneous wound healing in mammals can be subdivided into three successive overlapping phases: acute inflammation, proliferation and granulation tissue formation, and tissue remodelling. During the early inflammatory phase, platelet aggregation and blood coagulation leads to the formation of a clot. The extravasational part of the blood clot provides a provisonal matrix for the migration of cells such as polymorphnuclear leucocytes, lymphocytes, monocytes, fibroblasts, endothelial cells and pericytes (Dyson, 1997). Platelets secrete PDGF, TGF-alpha, and TGF-beta, which promote tissue generation, while pro-inflammatory agents enhance recruitment of leukocytes such as granulocytes and macrophages to the wound (Wolpe and Cerami, 1989).

During granulation tissue formation, new blood vessels grow into the wound bed. Granulation tissue is a well-vascularised, soft connective tissue in which the key cells are macrophages, fibroblasts and endotheliocytes. Cytokines derived from macrophages and fibroblasts, such as IL-1α, IL-1β, IL-8 and TNF-α orchestrate the formation of a fibroblast matrix, which replaces the provisional matrix provided by the blood clot (Hubner et al., 1996, Schaffer and Barbul, 1998, Agaiby and Dyson, 1999). In parallel, wound contraction, re-epithelisation, and angiogenesis occur. The remodelling phase of wound repair is characterised by collagen fibrillogenesis and development of extracellular matrix. The cellularity and vascularity of the reparative tissue decreases until finally the granulation tissue is replaced by scar tissue (Dyson, 1997).

Recent studies have confirmed a negative effect of stress on the rate of wound healing in animals (Padgett et al., 1998) and humans. One human study examined levels of chronic stress and wound healing in a group of Alzheimer’s caregivers and compared them with matched controls (Kiecolt-Glaser et al., 1995). Carers who exhibited high scores on the Perceived Stress Scale (PSS) showed slower wound healing rates and this difference was particularly marked at 14 days after a standardised punch biopsy. Likewise, in a sample of elderly people, individuals scoring high on the Hospital Anxiety and Depression Scale (HADS) exhibited significantly impaired healing of chronic wounds (Cole-King and Harding, 2001). Another study reported an intra-individual comparison of medical students during a phase of high psychosocial stress (i.e. during the exam period) versus a phase of relatively low stress (i.e. holiday period) (Marucha et al., 1998). In this study, standardised wounds to the dental palate during a holiday period healed significantly faster than equivalent wounds incurred during an exam period.

While these findings provide evidence for a clear relationship between heightened perceived stress and delayed wound healing in humans, some putative mediating factors have not been taken into account in previous studies. For example, health-risk behaviours such as smoking, alcohol consumption, poor sleep and lack of physical exercise have been associated with psychological distress (Hellerstedt and Jeffery, 1997, Baum and Posluszny, 1999, Vitaliano et al., 2002). The wound healing progress is highly dependent on the host’s nutritional status; especially glucose, polyunsaturated fatty acids, protein, and the vitamins A, C, E, and Zinc are essential dietary components during the healing progress (Russell, 2001, Scholl and Langkamp-Henken, 2001). Therefore, a lack of intake of these substances due to unhealthy eating habits (i.e. high saturated fat/low protein and vitamins), or vitamin depletion due to increased smoking and alcohol consumption could compromise wound healing (van den Berg et al., 2002). Additionally, disturbed sleep patterns due to stress could result in reduced growth hormone release and further downregulation of tissue repair processes (Lee and Stotts, 1990, Rose et al., 2001). Therefore, it is possible that the reported correlations between psychological distress and impaired wound healing could be secondary to stress-induced changes in health behaviours, which suppress the individual’s immune functions, thus delaying the repair process of the wound.

Most of the previous studies involved the use of macrophotography and hydrogen peroxide foaming to assess wound healing. However, hydrogen peroxide interacts with non-epithelial tissues causing tissue damage, and macrophotography only records the surface appearance of the wound. It has recently been demonstrated, that high resolution ultrasound (HRUS) scanning of a standard punch biopsy wound is a more valid measure of healing activity in deeper tissue layers than surface photography (Dyson et al., in press). Measures obtained by photography of the wound diameter were influenced by variable contractions of the wound scab, and especially by increases in wound diameter after the scab detaches. In contrast, non-invasive HRUS scans obtained at the base (i.e. at the level of the dermal/hypodermal junction) of a standard punch biopsy wound yielded measures documenting a more stable wound healing progress unaffected by changes in surface contractions (Dyson et al., in press).

Another putative mediating factor between psychological stress and wound healing are glucocorticoid levels. Glucocorticoid-induced suppression of the wound healing process has been documented in humans and animals (Goforth and Gudas, 1980, Gupta et al., 1999). This downregulating effect might be due to inhibition of cytokines such as IL-1, IL-6, IL-8 and TNF-α (Sapolsky et al., 2000) or growth factors, such as keratinocyte growth factor (KGF) 1 (Brauchle et al., 1995, Chedid et al., 1996), all of which play a pivotal role during the inflammatory-, re-epithelisation-, and fibroblast matrix-formation phase of wound regeneration (Dyson, 1997, Schaffer and Barbul, 1998). This hypothesis is supported by animal data demonstrating that glucocorticoid-induced immunosuppression leads to impaired dermal wound healing (Gupta et al., 1999). Studies in humans have shown that IL-1β levels were downregulated in individuals who exhibited slow wound healing (Kiecolt-Glaser et al., 1995, Marucha et al., 1998). Further, blister chamber fluid levels of the cytokines IL-1α and IL-8 were lower in participants with high perceived stress levels, and participants with the lowest levels of both cytokines at the induced blister wound site showed elevated cortisol levels in saliva (Glaser et al., 1999).

The cortisol response to awakening has been established as a reliable and stable marker of HPA axis activity in humans (Pruessner et al., 1997). Since this response has been found to be associated with levels of reported psychological distress (Schulz et al., 1998, Steptoe et al., 2000, Wüst et al., 2000), we investigated cortisol levels directly after awakening in relation to wound healing speed and levels of perceived stress.

In sum, the present study aimed to confirm and extend previous reports on positive associations between stress on wound healing. The effects of recent life stress during the preceding month on wound healing in a young adult sample were investigated using high resolution ultrasound (HRUS) scans to assess wound healing, controlling for health behaviours and cortisol levels. The goal of the study was to examine the relationship between perceived stress, emotional distress and wound healing over 21 days, using a longitudinal within-groups design.

Section snippets

Participants

Participants consisted of staff and students recruited via circular e-mails and posters at Kings College London. The mean age of the 24 participants was 29.42 (minimum 19, maximum 59, standard deviation 11.53 years). To minimise interindividual variation in cortisol levels due to gender (Kirschbaum et al., 1992), only males were recruited. Participants were screened for the following exclusion criteria via short interviews: Smoking, intake of glucocorticoid medication during the last month,

Results

All participants showed a significant progression of wound healing over the time between day 7 and day 21 after the biopsy. Average diameters of the wound base changed from 4.38 mm (std. dev. 0.58 mm) on day 7 to 3.52 mm (std. dev. 0.60 mm) on day 14 and to 2.80 mm (std. dev. 0.74 mm) on day 21. A repeated measures ANOVA confirmed a highly significant within subjects reduction in wound base diameter over time (F=49.9, p<.001).

Table 1 summarises the correlations between speed of wound healing

Discussion

In the sample reported here, wound healing assessed by repeated ultrasound scans was negatively correlated with perceived stress measured by the PSS and the GHQ, and positively with dispositional optimism. Further, the cortisol response in the morning of the day after the biopsy was negatively correlated with speed of wound healing. However, only inconsistent associations were observed between levels of cortisol after awakening and levels of perceived stress. Similarly, various health

Acknowledgements

We would like to thank Paul Wilson from Longport Intl. for technical support and scientific advice regarding the use of the portable ultrasound scanner.

References (49)

  • J.D Whitney et al.

    Modifying perfusion, nutrition, and stress to promote wound healing in patients with acute wounds

    Heart Lung

    (1999)
  • S Wüst et al.

    Genetic factors, perceived chronic stress, and the free cortisol response to awakening

    Psychoneuroendocrinology

    (2000)
  • A.D Agaiby et al.

    Immuno-inflammatory cell dynamics during cutaneous wound healing

    J. Anat.

    (1999)
  • A Baum et al.

    Health psychology: mapping biobehavioral contributions to health and illness

    Annu. Rev. Psychol.

    (1999)
  • M Calvin et al.

    Pilot study using high-frequency diagnostic ultrasound to assess surgical wounds in renal transplant patients

    Skin. Res. Technol.

    (1997)
  • M Chedid et al.

    Glucocorticoids inhibit keratinocyte growth factor production in primary dermal fibroblasts

    Endocrinology

    (1996)
  • S Cohen et al.

    A global measure of perceived stress

    J. Health. Soc. Behav.

    (1983)
  • A Cole-King et al.

    Psychological factors and delayed healing in chronic wounds

    Psychosom. Med.

    (2001)
  • M Dyson

    Advances in wound healing physiology: the comparative perspective

    Vet. Dermatol.

    (1997)
  • Dyson, M., Moodley, S., Verjee, L., Verling, W., Weinman, J., Wilson, P. Wound healing assessment using 20MHz...
  • T Field et al.

    Burn injuries benefit from massage therapy

    J. Burn Care Rehabil.

    (1998)
  • R.M Gallucci et al.

    Impaired cutaneous wound healing in interleukin-6-deficient and immunosuppressed mice

    FASEB J

    (2000)
  • R Glaser et al.

    Stress-related changes in proinflammatory cytokine production in wounds

    Arch. Gen. Psychiatry

    (1999)
  • P Goforth et al.

    Effects of steroids on wound healing: a review of the literature

    J. Foot Surg.

    (1980)
  • Cited by (254)

    View all citing articles on Scopus
    View full text