Introduction
Malignant wounds occur when cancerous cells invade the epithelium, infiltrate the supporting blood and lymph vessels, and penetrate the epidermis (O’Brien Reference O’Brien2012). This results in a loss of vascularity, and therefore, the loss of nourishment to the skin, leading to tissue death and necrosis (O’Brien Reference O’Brien2012). Malignant wounds are caused by tumor cell metastasis from distant primary tumors, primary cutaneous tumor, and direct invasion of a primary tumor into the cutaneous structures (Alexander Reference Alexander2009). Malignant wounds are found in 5–15% of all cancers (Beers Reference Beers2019; Tsichlakidou et al. Reference Tsichlakidou, Govina and Vasilopoulos2019). Patients with breast cancer have the highest prevalence of malignant wounds (Maida et al. Reference Maida, Ennis and Kuziemsky2009). As the mammary glands and lymph nodes are located directly under the skin, the cancerous cells can easily invade the epithelium (O’Brien Reference O’Brien2012).
The main symptoms that patients with cancer experience are fatigue, loss of appetite, pain, and dyspnea (Tewes et al. Reference Tewes, Baumann and Teufel2021), whereas the symptoms due to malignant wounds are more specific such as exudate, bleeding, odor, pruritus, and appearance changes. A previous report showed that 67.7% of patients with malignant wounds had at least one of these symptoms (Maida et al. Reference Maida, Ennis and Kuziemsky2009). In addition to these physical symptoms, patients with malignant wounds experience psychological symptom burden (O’Brien Reference O’Brien2012). Malignant wounds are lesions on the body surface that are directly visible to the patients and are, thus, a visible reminder for the patients that their body is breaking down in the dying process. Thus, patients with malignant wounds suffer from a devastating and often crippling symptom burden (Tilley et al. Reference Tilley, Lipson and Ramos2016). Hence, psychological support, as well as palliation of physical symptoms, is required to maintain their quality of life (QOL).
Despite the physical impact of malignant wounds being reported (Maida et al. Reference Maida, Ennis and Kuziemsky2009), the psychological impact of malignant wounds on QOL and their end-of-life status are poorly investigated. We hypothesized that patients with malignant wounds suffer from a psychological symptom burden due to a visible reminder that their body is breaking down in the dying process. Therefore, this study evaluates and clarifies the psychological distress associated with malignant wounds among patients at their end of life.
Methods
Study cohort
This study conducted a secondary analysis of the results of a large prospective cohort study known as the East Asian Collaborative Cross-cultural Study to Elucidate the Dying Process, which investigated the dying process among patients with advanced cancer in 23 palliative care units (PCUs) in Japan, between January 2017 and June 2018 (Hiratsuka et al. Reference Hiratsuka, Yamaguchi and Maeda2020). The inclusion criteria of the present study were (1) adult patients (18 years of age or older), (2) patients diagnosed with locally advanced or metastatic cancer, and (3) patients admitted to PCUs. Patients with missing data for outcome variables were excluded from the present study.
The protocol of this study was approved by the Institutional Review Board of each participating center, and informed consent was waived in accordance with local regulations. Patient registration was based on an opt-out model. This study was conducted according to the principles of the Declaration of Helsinki.
Procedure
In this secondary analysis, patients’ clinical characteristics (i.e., sex, age, primary cancer site, presence of metastasis, and the Eastern Cooperative Oncology Group-Performance Status [ECOG-PS]), information about medication use, specific symptoms due to malignant wounds (bleeding, exudate, odor, and itching), and the status of malignant wounds were obtained upon admission. The physical and psychological symptom burden of patients was evaluated by the Integrated Palliative Care Outcome Scale (IPOS). The frequency of dressing changes was recorded as well.
IPOS
Physicians in the PCUs rated patients’ physical symptom burden upon their admission on the 5-point Likert-IPOS, ranging from 0 = not at all; 1 = slightly; 2 = moderately; 3 = severely; to 4 = overwhelmingly (Hearn and Higginson Reference Hearn and Higginson1999). The rated physical symptoms were pain, breathlessness, fatigue, anorexia, somnolence, and dry mouth. The IPOS is a valid and reliable outcome measure, which can assess and monitor symptoms and concerns in advanced illnesses, determine the impact of health-care interventions, and demonstrate the quality of care (Murtagh et al. Reference Murtagh, Ramsenthaler and Firth2019).
The psychological symptom burden was evaluated by the “IPOS-feeling at peace” score. IPOS-feeling at peace is a score in which health-care professionals rate patients’ emotional symptoms regarding whether she/he has felt at peace, ranging from 0 = always; 1 = most of the time; 2 = sometimes; 3 = occasionally; to 4 = not at all (Hearn and Higginson Reference Hearn and Higginson1999). The inter-rater agreement of IPOS-feeling at peace between the patient and staff ratings was 72.4% (Murtagh et al. Reference Murtagh, Ramsenthaler and Firth2019).
The primary outcome of this study was the prevalence of moderate to severe psychological symptom burden, evaluated as an IPOS-feeling at peace scores of 2–4. Subsequently, the factors affecting psychological symptoms were also investigated.
Numerical rating scale for pain
Patients’ pain was assessed using the numerical rating scale for pain (Pain NRS) in addition to the IPOS-pain score. Pain NRS is a scale to evaluate pain by the patients themselves; it is a single 11-point numeric scale (ranging from 0 to 10 from no pain to worst possible pain).
Good death scale
The quality of death of patients was evaluated upon death using the Japanese version of the Good Death Scale (Kodama et al. Reference Kodama, Kobayashi and Katayama2017). The Good Death Scale consists of 5 domains: (1) awareness; awareness about one’s dying (0 = complete ignorance, 1 = ignorance, 2 = partial awareness, and 3 = complete awareness), (2) acceptance; peaceful acceptance of death (0 = complete unacceptance, 1 = unacceptance, 2 = acceptance, and 3 = complete acceptance), (3) propriety; honoring the patient’s wishes (0 = no reference to the patient’s wishes, 1 = following the family’s wishes alone, 2 = following the patient’s wishes alone, and 3 = following the wishes of the patient and the family), (4) timeliness; death timing (0 = no preparation, 1 = the family alone had prepared, 2 = the patient alone had prepared, and 3 = both the patient and the family had prepared), (5) comfort; the degree of physical comfort 3 days before death (0 = a lot of suffering, 1 = suffering, 2 = a little suffering, and 3 = no suffering) (Higashibata et al. Reference Higashibata, Hamano and Hisanaga2022). The outcome included the score for each domain and its total score. The Good Death Scale was developed as a quality-of-death measurement instrument for patients receiving palliative care in a hospice or a hospital and was completed by health-care providers. Its internal consistency, reliability, and validity have been confirmed for the Asian culture (Higashibata et al. Reference Higashibata, Hamano and Hisanaga2022). The validity of a proxy evaluation of the Good Death Scale score, which is evaluated by a physician instead of the primary caregiver, has been reported (Cheng et al. Reference Cheng, Dy and Huan2013).
Statistical analyses
Comparisons among the 2 groups (patients with malignant wounds vs. patients without malignant wounds; patients whose IPOS-feeling at peace score 0–1 vs. 2–4) were made using the Fisher’s exact test or Student’s t-test as appropriate. Univariate and multivariate logistic regression analyses were also performed to evaluate the patient characteristics (age, sex, IPOS-pain, and IPOS-feeling at peace) associated with malignant wounds. All statistical analyses were performed using the JMP software (version 14.3.0 for Windows; SAS Institute Japan Inc., Cary, NC, USA). The results were considered statistically significant at a two-sided p-value of <0.05.
Results
Patient characteristics
From January 2017 to June 2018, 1896 patients were admitted to PCUs. Among them, 965 (50.9%) were males and 931 (49.1%) were females. The mean age was 72.4 years (range: 25–100). The major primary site was gastrointestinal in 519 patients (27.4%), hepatobiliary and pancreatic in 363 patients (19.1%), and lung in 319 patients (16.8%). The ECOG-PS of more than 90% of patients was 3 or 4.
Among 1896 patients, 156 (8.2%) had malignant wounds and 1740 (91.8%) did not have malignant wounds. Malignant wounds predominantly affected females (p = 0.001) and younger patients (p = 0.004). Breast (44 patients, 28.2%) and head and neck (30 patients, 19.2%) were the most prevalent primary sites for patients with malignant wounds. Table 1 presents the characteristics of the patients.
Patient characteristics
ECOG: Eastern Cooperative Oncology Group; MW: malignant wound; PS: performance status.
The site and condition of malignant wounds
Malignant wounds occurred mainly on the chest and abdomen (n = 75, 48.1%) and face (n = 30, 19.2%). Skin conditions observed due to malignant wounds were as follows: reddening (n = 59, 37.8%), skin defects (n = 38, 24.4%), necrosis (n = 50, 32.1%), and fistula (n = 24, 15.4%). On the one hand, the frequency of dressing changes was once a day or never in 78.8% of patients (n = 123) with malignant wounds. On the other hand, 21.2% of patients (n = 33) required dressing changes twice or more a day (Table 2).
The site, condition, and specific symptoms of malignant wounds
MW: malignant wound.
a Fourteen patients had malignant wounds with multiple sites.
b Fifteen patients had malignant wounds with multiple conditions.
c Specific symptoms of MWs: exudate, bleeding, odor, and itching.
Specific symptom prevalence due to malignant wounds
Among 156 patients with malignant wounds, exudate, bleeding, odor, and itching were specific symptoms due to malignant wounds and were observed in 77 (49.4%), 43 (27.6%), 29 (18.6%), and 7 (4.5%) patients, respectively. Among them, 44 patients (28.2%) had multiple symptoms due to malignant wounds (Table 2).
Physical and psychological symptom burden of the patients on admission
On admission, significant differences were found in the prevalence of moderate to severe IPOS-pain score (IPOS 2–4) between patients with or without malignant wounds (44.9% vs. 34.1%, p = 0.013), while the median Pain NRS was both 2.0 for patients with and without malignant wounds (p = 0.41). More patients with malignant wounds used opioids than patients without malignant wounds (76.3% vs. 62.5%, p = 0.001) and higher doses (mean oral morphine equivalent daily dose: 87.1 vs. 66.3 mg, p = 0.03). Results showed that 64 (41.0%) patients with malignant wounds and 544 (31.3%) patients without malignant wounds had IPOS-feeling at peace scores of 2–4, respectively (p = 0.024). More patients with malignant wounds used antidepressants than patients without malignant wounds (7.7% vs. 3.7%, p = 0.03) (Table 3).
Symptom burden of the patients with and without MWs on admission
IPOS: Integrated Palliative Care Outcome Scale; MW: malignant wound; NRS: numerical rating scale; OMEDD: oral morphine equivalent daily dose.
a Psychotropic drug includes antipsychotics, benzodiazepine, non-benzodiazepine hypnotic, or antidepressant.
Multivariate analysis showed that malignant wounds were significantly more common in females (p = 0.001). Patients with malignant wounds tended to have moderate to severe IPOS-pain scores (p = 0.10) and IPOS-feeling at peace scores (p = 0.072), however, not significantly (Table 4).
Univariate and multivariate analyses of patient characteristics associated with malignant wounds
CI: confidence interval; IPOS: Integrated Palliative Care Outcome Scale.
The factors affecting psychological symptoms
Among 156 patients with malignant wounds, 149 patients were evaluated for their psychological symptom burden using IPOS-feeling at peace scores (Fig. 1). IPOS-feeling at peace score was associated with the IPOS-pain score and the frequency of dressing changes. Among the 79 patients who had an IPOS-pain score of 0–1 and 69 patients who had an IPOS-pain score of 2–4, 26 (32.9%) and 38 (55.1%) patients had an IPOS-feeling at peace scores of 2–4, respectively (p = 0.008). Among the 32 patients with malignant wounds who required dressing change at least twice a day and 117 patients with malignant wounds who required dressing change once a day or never, 20 (62.5%) and 44 (37.6%) patients had an IPOS-feeling at peace scores of 2–4, respectively (p = 0.015).
Flow diagram of patient selection.
On the other hand, IPOS-feeling at peace score was not associated with the presence of malignant wounds on the face, condition of malignant wounds, or specific symptoms due to malignant wounds (bleeding, exudate, odor, and itching), oral morphine equivalent daily dose, antidepressant use, and the number of symptoms (Table 5).
The factors affecting psychological symptoms for patients with MWs
IPOS: Integrated Palliative Care Outcome Scale; MW: malignant wound, OMEDD: oral morphine equivalent daily dose.
Good Death Scale scores in patients with and without malignant wounds
There was no difference between the patients with or without malignant wounds in awareness, acceptance, timeliness, comfort, and the total score on Good Death Scale, whereas the propriety score was significantly higher among patients with malignant wounds than among patients without malignant wounds (p = 0.019). The Good Death Scale scores are shown in Table 6.
Good Death Scale scores for the patients with and without MWs
MW: malignant wound.
Discussion
Among patients with advanced cancer at their end of life, 8.2% of patients developed malignant wounds. Malignant wounds were more common in female and young people, probably due to the high prevalence of breast cancer. Multivariate analysis also showed that malignant wounds were significantly more common in females. Patients with malignant wounds frequently required antidepressants, which might be associated with psychological distress due to malignant wounds. Patients with malignant wounds had higher IPOS-feeling at peace scores, suggesting higher psychological distress, which was associated with moderate to severe IPOS-pain and the frequency of dressing changes due to exudate and bleeding.
Patients with malignant wounds had higher IPOS-pain scores and required higher doses of opioids. This suggests that they have more pain than patients without malignant wounds. At the same time, they did not have higher Pain NRS than patients without malignant wounds. Pain NRS is a score to evaluate pain by the patients themselves. IPOS is a scale where health-care professionals rate the patient’s symptoms. Discrepancies between the health-care professionals’ assessment and the patient’s complaints of pain might exist. A previous report showed that there was no significant correlation between Pain NRS and IPOS (Amano et al. Reference Amano, Ishiki and Miura2021). The influence of opioids also cannot be ignored. Pain NRS for patients whose pain is controlled by high-dose opioids may be underestimated when compared to the IPOS-pain score, which is rated by health-care professionals.
Patients with malignant wounds tended to have higher IPOS-feeling at peace scores and IPOS-pain scores, however, not significantly. Presence of confounding variables were suspected. Breast cancer accounts for the majority of patients with malignant wounds. Many breast cancer patients have reported that they often experience multiple concurrent psychological symptom burden due to major changes in appearance by cancer development and mastectomy (Guimond et al. Reference Guimond, Ivers and Savard2020; Izydorczyk et al. Reference Izydorczyk, Kwapniewska and Lizinczyk2018), which might result in higher IPOS-feeling at peace scores and IPOS-pain scores.
Psychological distress was associated with the frequency of dressing changes. Malignant wounds which need frequent dressing changes are advanced and cause many physical symptoms, which adds to the psychological distress. In addition to physical symptoms due to malignant wounds, the procedure of dressing changes itself can contribute to psychological distress for them because of the pain due to the procedure and the time it takes. Patients may feel apologetic toward the caregiver as they have to burden them with changing the dressing. Through the procedure of dressing changes, patients might realize that their body is breaking down, which thus, might lead to psychological distress.
To maintain their QOL, adequate pain control and less uncomfortable dressing changes will be required. Also, health-care providers need to understand that patients suffer from psychological distress by these factors.
The presence of malignant wounds did not affect the quality of death. Only the “propriety” score was significantly higher among patients with malignant wounds than patients without malignant wounds, meaning that the wishes of patients with malignant wounds and their families were more respected. Patients with malignant wounds had more opportunities to undergo procedures and might feel that their wishes were respected. On the other hand, the presence of malignant wounds did not affect the Good Death Scale total score. The quality of death is determined by numerous factors that influence its judgment, including culture, type and stage of disease, and social and professional roles in the dying experience (Hales et al. Reference Hales, Zimmermann and Rodin2008). Therefore, the presence of malignant wounds alone might not affect the QOL total score.
This study has several strengths. First, this is a large, prospective, and multicenter study. While a previous report included 67 patients with malignant wounds (Maida et al. Reference Maida, Ennis and Kuziemsky2009), we included more than 150 patients with malignant wounds, out of nearly 2000 patients with cancer. Additionally, a range of information was obtained, including clinical characteristics, medications, symptom burden, and the malignant wound status. Finally, the large number of data obtained and the small number of missing data provide a high degree of generalizability.
This study has several limitations. This is a post hoc analysis of a prospective cohort, and all data were obtained from Japanese patients in PCUs. Although this study showed the relationship between pain and psychological distress in patients with malignant wounds, this relationship may not be limited to patients with malignant wounds. The psychological distress of non-end-of-life population with malignant wounds may be different. In addition, patients’ feelings may have been impacted by the negative emotion experienced during admission to PCUs, and thus, the evaluation may not be accurate. Feelings and symptoms that could have changed after hospitalization were not evaluated. Moreover, IPOS and Good Death Scale were evaluated by health-care professionals and not based on patient-reported outcomes. Although the validity of a proxy evaluation has been reported (Cheng et al. Reference Cheng, Dy and Huan2013; Murtagh et al. Reference Murtagh, Ramsenthaler and Firth2019), the results of the present study should be interpreted with caution. Studies based on data on patient-reported outcomes are necessary. At the same time, evaluating the symptom of patients at end of life using the patient-reported outcomes are limited because of their poor general condition and the difficulty of accurately describing their own symptom burden.
Therefore, despite the limitations, this study is a worthy report, which showed the psychological impact of malignant wounds on patients.
Conclusions
This study showed that patients at their end of life suffer from psychological distress due to malignant wounds. Psychological distress was associated with moderate to severe IPOS-pain and the frequency of dressing changes. Adequate pain control and less uncomfortable dressing changes will be required. Patients with malignant wounds require psychological support in addition to the treatment of physical symptoms for maintaining their QOL. Also, health-care providers need to understand their psychological symptom burden.
Acknowledgments
We thank all the patients whose data were used for the study. Editage (Cactus Communications) provided editorial support in the form of medical writing, table assembly, collating author comments, copyediting, fact-checking, and referencing based on the authors’ detailed directions.
Funding
The present study was supported in part by a grant-in-aid from the Japan Hospice Palliative Care Foundation.
Competing interests
Dr. Ishiki reports personal fees from Mundipharma, DaiichiSankyo, Shionogi, and EA Pharma outside of the submitted work. All other authors state that they have no conflict of interest.
