Prevalence of Pressure Injury and Associated Factors in Hospitalized Adult Patients With Cancer

Hospitalized cancer patients experience limitations in their daily activities resulting from the severity of their condition. Such changes may alter their sensory perception and entail loss of mobility, predisposing them to PI, negatively affecting quality of care, and placing them at increased risk of additional complications, which in turn influence patient morbimortality, with prolonged length of stay and increased treatment costs.^1,2

Critically ill patients have increased clinical needs and require complex care owing to hemodynamic instability, and they are at high risk of developing PI, among other complications.^3,4

The prevalence and incidence of PI are associated with increased morbidity and mortality. A prospective worldwide observational 1-day point prevalence study of PIs among adult ICU patients with a maximum 12-week follow-up period for survival status and length of hospital stay identified 26.6% overall prevalence of PI among ICU patients.⁵ Considerable regional variation in prevalence was noted. Approximately 60% of patients developed PI in the ICU, regardless of regional prevalence. In the United States, PIs of the skin and soft tissues affect up to an estimated 3 million people each year.⁶ The incidence differs based on the clinical setting. Among hospitalized patients, the reported prevalence rates were 5% to 15% of patients overall.⁶ The percentage was considerably higher in patients in palliative home care (range, 18.8%⁷–26.5%⁸) and in ICUs (26.6%).⁵ In Canada, the estimated prevalence of PI ranges from 23.9% to 29.7% among acute care patients.⁹

In a retrospective cohort of 766 patients in 9 ICUs of 2 university hospitals in Brazil, the PI incidence was 18.7%. The OR for developing PI was 3.5 times greater in patients on mechanical ventilation (P <.001), 7.8 times greater in patients in palliative care (P =.004), and 2.3 times greater in patients aged 60 years to 84 years (P =.005).¹⁰ Additionally, it increased 10% for each day of hospitalization (P <.001) and 1.5% for each additional point of the Nursing Activities Score (P =.016).

A prospective, descriptive, exploratory study evaluating the implementation of a pressure prevention protocol among patients in the ICU setting reported a PI incidence of 23.1%.¹¹ This was lower than the rate of 41.02% reported in a similar study at the same institution before the implementation of risk assessment and PI prevention protocols.¹²

Although terminally ill patients are at high risk of PI because of their more vulnerable global condition, sparse evidence exists about the prevalence and incidence of PI in this patient population. A study conducted in the Toronto region (Canada) recruited patients through a palliative program who were followed until their death or discharge from the program.¹³ Of the 593 patients eligible for inclusion in the study, 9 wound type categories were created, including PI; 60.6% of patients with and without cancer had stage I, II, III, IV, or unstageable PI according to the NPUAP/EPUAP classification system. The prevalence of PI in cancer patients was 22.4 (n = 93), and the incidence of PI (incidence rate per month per 100 patients with cancer) was also 22.4 (n = 182). This prospective observational study based on a sequential case series of 593 patients with advanced illness referred for supportive and palliative care demonstrates that wounds represent a major domain of care within the overall medical and nursing management of these patients. PIs constituted the most commonly occurring wound class, affecting more than 50% of all patients.

Jomar et al¹⁴ reported a 29.5% PI incidence in an oncology ICU with data collected from May 2017 to November 2017. Some studies examining PI prevalence rates in terminally ill patients receiving palliative care, including patients with cancer, report rates of 13.1% to 26.57%.^8,15,16 Hendrichova et al¹⁶ reviewed the medical records of 414 patients with cancer who were receiving palliative care and reported a PI prevalence rate of 22.9%.

In an Italian multicenter prospective cohort study that enrolled 440 patients with advanced-stage cancer who were consecutively admitted to 5 hospices during a 1-year period, the incidence of PIs in the total population was 17.3%.¹⁷ The risk factors that influenced the development of PIs were age, proximity to death, and duration of stay in hospice.

The literature is sparse regarding PI in cancer patients in hospital settings. The goal of this study was to analyze the prevalence of PI and associated factors in such patients.

This study is part of an original research study analyzing adult inpatients and ICU patients treated at an oncological hospital in São Paulo, Brazil. This observational, epidemiological, cross-sectional, descriptive study is being conducted at a nonprofit institution that provides care for patients with private health insurance and those who are insured through the National Health Care System. The research population consists of all adult patients admitted between November 23, 2015, and December 1, 2015. The facility is composed of 3 towers, with 385 total adult beds, on average, without subspecialty units’ distribution.

The sample included all hospitalized adult patients with cancer admitted for adult inpatient treatment or to an ICU who agreed to participate and who signed the Informed Consent Form. A caregiver signature was requested for patients who were unable to provide interview responses. It is important to note that patient inclusion in the sample did not take into account length of stay, and no patient was assessed twice.

To meet the study goals and because it would have been impossible to collect data from all patients hospitalized at the institution in active beds in a single day, and because it was necessary to avoid counting eligible patients more than once, patients in each tower were assessed sequentially over 9 consecutive data collection days. Patients were interviewed to obtain social and demographic information (age, sex, race, and marital status) and clinical data (diagnosis at hospitalization, previous and current treatment, alcoholism and smoking) alterations of painful and tactile sensitivity, alteration in the pattern of sleep and rest, emotional stress, gait, presence of bruise, edema in the lower and upper limbs, body mass index, use of medications, presence of infection, presence of pain). Clinical data were also obtained based on general physical examination consisting of visual inspection and skin palpation to assess for the presence of bruising, hematoma, and dry and scaly skin; senile purpura; upper limb and lower limb edema; use of support equipment; use of orthopedic shoes; use of catheter drainage; use of disposable diapers; and skin adhesive tapes and dressings. Additional information was complemented or provided by caregivers and medical records. After general physical examination, the patients also underwent dynamic examination to assess physical mobility. For this assessment, patients were observed for a few minutes while they walked a straight line down the ward. For ICU patients, physical mobility was assessed using the Braden Scale.

Upon identification, each PI was assessed to determine and record stage, number, and location according the NPUAP/EPUAP classification system published by the NPUAP, EPUAP, and PPPIA in 2014.¹⁸

Results are presented as relative and absolute frequencies for sample characterization. Associations between the dependent variable (presence of PI) and independent variables were first obtained using the chi-square test and by calculating the OR and associated 95% CI in the univariate analysis. The forward stepwise (likelihood ratio) method was used for logistic regression analysis. Tests that obtained an α level of .05 were considered statistically significant for the remaining analyses.¹⁹

Point prevalence is the ratio of individuals who present with a clinical condition at a certain point in time.¹⁹ It was calculated based on the following formula: Prevalence = (Nº of patients with PI)/(total Nº of patients × 100), where Nº = number.¹⁹

The research protocol was approved by the Fundação Antônio Prudente Research Ethics Committee (CEP nº2088/15, CAAE: 46697115.5.0000.5432), according to Brazilian Ministry of Health Resolution Nº 466/12 and international ethical laws.²⁰

Of the 369 cancer patients recruited (16 empty beds), 28 (7.6%) were excluded from this study: 1 patient was hemodynamically unstable, 2 patients in the ward had previously been assessed in the ICU, and 25 patients (6.8%) refused to participate in the study. Thus, a total of 341 (92.4%) patients comprised the final sample size.

Of the 341 patients remaining the PI prevalence was 10% (34 patients), with an average age of 61.5 years in that smaller group (SD = 15.6). Of those 34 patients, 58.8% were 60 years and older, and of that group, 14.7% were 80 years and older. There was predominance of male patients (58.8%) and White patients (50.0%) in the group with PI. For 41% of these patients, race was not indicated in their medical records. Most patients were married (53.0%). Most patients (62.0%) used disposable diapers, 13.0% had an indwelling urinary catheter, and 14.0% received diuretics.

The most common location of PI was the sacrum (64%, 22 patients), followed by the calcaneus (15.0%, 5 patients) and the buttocks (9%, 3 patients) (Figure 1). Most PIs were classified as stage II (41.2%, 14 patients) or stage I (33.0%, 11 patients) (Figure 2).

The presence of bruising (P = .010), hematoma (P = .028), and upper limb edema (P = .028), as well as the use of diapers (P = .001) was more frequent in patients with PI than in those without PI. Although bruising and hematomas are present, these were not located in the main regions where pressure injuries were identified–sacral, calcaneus, and buttocks. Pain was reported by 38.2% of patients during dressing change, with 30.7% at dressing removal. Continuous pain was reported by 23.0% of patients, and 7.7% reported pain when changing bed position.

According to logistic regression analysis, only the use of disposable diapers was related to the development of PI in hospitalized cancer patients, with such patients 6-fold increased chance of developing PI (OR, 6.077; 95% CI, 2.884–12.806; P =.05).

Few epidemiological studies on PI among patients with cancer have been published in Brazil or elsewhere. The PI prevalence of 10% found in the present study is even lower than that reported in a Brazilian study published in 2013 of PI and nutritional status in hospitalized patients in various regions of the country.²¹ In that earlier study, 23.5% of patients had a cancer diagnosis, and the global prevalence of PI was 16.9%. Sankaran et al²² reported a PI prevalence of 19.4% among patients with terminal cancer. Healing of PI occurred in only 9 patients (42.9%), and the median time to healing was 56 days. Jakobsen et al¹⁷ reported a 25.5% prevalence of PI among cancer patients on admittance to hospice (point prevalence). The period prevalence, however, was 42.8% (188 of 440), and the incidence of PI in the total population was 17.3%.

The sacral region was the anatomic location most affected by PI in the present study, which is like other studies.^7,23–25 This may be because the supine position is the most common for hospitalized patients. In the current study, stage I and stage II PIs were the most prevalent. This differs from the findings of a 2014 study⁷ in which patients with cancer had 1 to 3 PIs for a total of 19 lesions; of those, 9 PIs (47.4%) were stage III. A higher incidence of stage III PI was also more prevalent in 9 patients (47.4%) under palliative care.⁷

In that study, the presence of PI was higher among those with a history of previous wound. PI was deemed a significant event in the study population, which suggests that the home palliative care team should incorporate preventive measures.

In the studied institution (AC Camargo Cancer Center), skin specialist nurses have developed a workflow focused on PI prevention through an evidence-based protocol, using technologies that help reduce PI incidence and promote early identification. This may be the reason less severe stages of PI were identified in the present study compared with other studies. Many studies have confirmed that PI prevention protocols significantly reduce the PI prevalence and incidence rates.^11,26,27

In the present study, patients using disposable diapers had a 6.077 increased chance of PI occurrence. The literature shows that excessive humidity and loss of natural skin oil owing to frequent cleansing in incontinent patients increases the risk of PI.^26,28,29 In a patient with a PI who is fitted with only a urinary catheter, whose incontinence pad remains dry, and who has only 1 incontinence pad, it must be assumed that the pad is the source of moisture.¹⁷ According to Engels et al²⁸, incontinent surgical patients have a 56-fold increased likelihood (P =.033) of developing PI compared with surgical patients who are not incontinent; these findings corroborate those of the current study. Most patients (62.0%) in the present study used disposable diapers; in addition, 59.0% were aged 65 and older, and some used diuretics.

Even with the publication of international clinical practice guidelines in 2019²⁶ and the implementation of the Brazilian Patient Safety Policies in 2014,³⁰ hospitalized cancer patients in Brazil experience increased morbidity and mortality associated with the occurrence of PI. Based on the current results, it is recommended that multidisciplinary teams work together to improve PI prevention guidelines, starting with recognizing the problem from calculating the prevalence and associated variables to know the population better and identify conditions that need special attention. As in this study, the use of diapers increased the development of PI. Therefore, reviewing and optimizing the device indication only during the patient’s real need is recommended.

The study setting is among one of the essential oncology reference centers in Brazil; however, the data collected were from a single institution. Other limitations included the sample size, study duration, the data (from 2015), and the characteristics of the population may have changed in the recent years.

In this epidemiological study of 341 hospitalized cancer patients, the prevalence of PI was 10% (34 patients). The use of disposable diapers was the only factor associated with the presence of PI after logistic regression analysis. It was essential to know the risk factors of this patient population to implement specific actions and optimize indications for disposable diaper use. Unfortunately, despite the current study, there are few publications on the topic, leading to an absence of epidemiological records. Therefore, it is important to develop a program for PI prevention in the inpatient, assisted patient population; educational programs for patients and clinicians, protocols, and monitoring of results may be beneficial.