Breakthrough pain among cancer patients at oncology units in Northern Ethiopia; a multi-center study

Background Breakthrough cancer pain (BTCP) is a transient exacerbation of pain that affects the length of hospitalization and quality of life of patients. The objective of this study was to determine the prevalence and factors associated with BTCP among cancer patients at oncology units in Northern Ethiopia in 2022. Methods A multi-center cross-sectional study was conducted from April to June 2022. After obtaining ethical approval, data were collected prospectively from 424 adult cancer patients admitted to oncology units. Breakthrough cancer pain was assessed by the numeric rating scale. Descriptive and binary logistic regression analyses were performed to determine the factors associated with BTCP. The strength of association was described in adjusted odds ratio (AOR) with 95% confidence intervals and variables with a P-value < 0.05 were considered to have a statistically significant association with BTCP. Result The prevalence of BTCP among cancer patients was 41.5%. The factors that were found to be associated with BTCP were colorectal cancer (AOR: 7.7, 95% CI: 1.8, 32.3), lung cancer (AOR: 6.9, 95% CI: 1.9, 26.0), metastasis (AOR: 9.3, 95% CI: 3.0, 29.1), mild background pain (AOR: 7.5, 95% CI: 2.5, 22.6), moderate background pain (AOR: 7.0, 95% CI: 2.2, 23.1), severe background pain (AOR: 7.1, 95% CI: 2.2, 22.8), no analgesics taken for background pain (AOR: 5.1, 95% CI: 2.8, 9.3) and uncontrolled background pain (AOR: 3.3, 95% CI: 1.8, 6.1). Conclusion The prevalence of BTCP was high. Colorectal cancer, lung cancer, the presence of metastasis, the presence of background pain, not taking analgesics for background pain, and uncontrolled background pain were significantly associated with BTCP.


Introduction
Cancer is a multi-symptomatic disease associated with physical, social, and emotional sequelae, and it is one of the leading causes of death worldwide (1).Overall, nearly 10 million deaths were reported in 2020.Data from the World Health Organization's Global Cancer Observatory (GLOBOCAN) showed that breast (2.26 million), lung (2.21 million), and colorectal (1.93 million) cancers were the most frequent new cases in 2020 (2).In sub-Saharan Africa in 2020, breast and cervical cancer were the most common, and prostate cancer was the leading incidence of cancer in men (3).
Cancer pain is one of the most complex symptoms due to its cognitive, behavioral, sensory, and emotional components.It is the most frightening symptom with a prevalence of 50% to 70%.The highest prevalence of pain is commonly observed among pancreatic and head/neck cancer (4)(5)(6)(7).
Breakthrough cancer pain (BTCP) is defined as a transient exacerbation of pain that occurs either spontaneously or in relation to a specific, predictable, or unpredictable trigger, despite having relatively stable and adequately controlled background pain (8).It is a common clinical condition in cancer patients characterized by rapid onset, short duration, and severe intensity.It has a significant impact on a patient's quality of life (9)(10)(11).Breakthrough cancer pain is categorized as spontaneous pain and incidental pain.Spontaneous pain is unpredictable and its precipitants are not clearly identifiable, whereas incidental pain is somewhat predictable (12)(13)(14).The consequences of poor breakthrough cancer pain management include poor functional ability, poor sleep patterns, increased anxiety and depression, poor compliance with cancer treatments, increased costs of care, and patient dissatisfaction (9,(15)(16)(17).A variety of factors such as sociodemographic factors, the type and site of cancer, stage, metastasis, physical activity, background pain, and pain medications have been found associated with BTCP (4,(18)(19)(20)(21)(22)(23)(24)(25)(26)(27)(28)(29).
Breakthrough cancer pain is commonly overlooked by healthcare providers.Despite its deleterious consequences, the management of BTCP is still sub-optimal (5).Limited researchbased evidence is available regarding BTCP in low-and middleincome countries, particularly in sub-Saharan Africa.The general objective of this study was to assess the prevalence and factors associated with BTCP at oncologic units in Northern Ethiopia.

Study populations, and inclusion and exclusion criteria
All adult (18+) cancer patients admitted to oncology units in the aforementioned hospitals in the Amhara National Regional State during the study period were included.Patients who were unable to communicate and volunteer to participate in the study were excluded.

Variables and operational definitions
The dependent variable was breakthrough cancer pain and the independent variables were socio-demographic characteristics (age, sex), cancer characteristics (type, site, stage, and metastases), treatment-related factors, and background pain intensity.
Breakthrough pain: a transitory increase in pain to a score greater than NRS ≥ 7/10, which occurred alongside or further to a background pain (30).
Background pain: a mild to moderate intensity of the pain with a value of 1 to 6 on NRS (31).
Controlled background pain: background pain that is managed with scheduled opioids and non-opioid analgesics and has an NRS of < 3/10 (32).
Numerical Rating Scale: A pain assessment tool in which a number is assigned from 0 -10 to represent the severity of the pain; 0 = no pain; 1 -3 = mild pain 4 -6 = moderate pain 7 -10 = severe pain (33).
Somatic pain: Pain arising from the skin, subcutaneous tissue, and musculoskeletal system characterized by being intense, constant, and localized to a particular area (34).
Visceral pain: Pain originating from internal organs characterized by a diffuse, dull, and poorly localized type of pain (35).
Neuropathic pain: Pain resulting from damage to the nervous system characterized by burning, electrical shocks, shooting, and needle-type pain (36,37).

Sample size determination and sampling technique
The sample size was determined by using a single population proportion formula with the assumptions of 50% prevalence rate and 5% margin of error at 95% confidence level.The final sample size was 424 with the addition of 10% non-response rates.All consecutive cancer patients who were admitted to those oncology units were included in this study center until the calculated sample size was attained.There were 2,736 cancer patients receiving cancer treatment follow-up in the three oncology units (864 at UoGCSH, 720 at FHCSH, and 1,152 at DCSH).The sample taken from each unit was calculated by using a proportion formula (134 UoGCSH, 112 FHCSH, and 178 from DCSH) (Figure 1).

Ethical approval, data collection procedure, quality control, and analysis
Ethical approval was obtained from the Ethical Review Committee of the School of Medicine, University of Gondar (Reference number: SOM/1404/2022), and permission to conduct data collection was obtained from each hospital.Written informed consent was obtained from each study participant after a detailed explanation of the study.All methods were performed in accordance with the relevant research guidelines and regulations.
The data were collected by three trained nurses by interviewing the participants and chart review, using a structured questionnaire.The questionnaire was adapted from the Alberta Breakthrough Pain Assessment.It was found valid and reliable to assess BTCP (Cronbach's alpha = 0.83) (19,32,38,39).To ensure the quality of the data, a pretest was conducted on 22 patients (5% of the sample).The accuracy, completeness, and consistency of the collected data were checked; and data collectors were closely supervised by the investigators.
The collected data were cleaned, coded, entered, and analyzed by SPSS version 26 (IBM Corporate).The normality of the data were tested using the Shapiro-Wilk normality test.Variables with variance inflation factors of less than 10 were considered to have no collinearity.The fitness of the model was checked using the Hosmer-Lemeshow goodness-of-fittest.Descriptive, Chi-square, and binary logistic regression analyses were employed to determine the factors associated with BTCP.Variables with a P value < 0.2 on bivariate logistic regression were entered into multivariate logistic regression analysis.In the multivariable logistic regression analysis, variables with a p-value of < 0.05 were considered to have a statistically significant association with BTCP.The strength of association was described in adjusted odds ratio (AOR) at a 95% confidence interval (CI).

Prevalence of breakthrough cancer pain
We found that 176 patients had BTCP (41.5%, 95% CI = 36.8-46.2).The prevalence of BTCP among female patients was 59.1%, whereas it was 40.9% among male patients.Among those 176 patients who had BTCP, 136 (77.3%) used medication to relieve BTCP, while the remaining 40 (22.7%)subjects did not use any medication to relieve their pain.The 56 (41.2%) patients were taking Morphine to relieve BTCP, while the remaining 42 (30.9%)and 38 (27.9%) took NSAID and NSAID with other combinations respectively (Table 2).The prevalence of BTCP was relatively high in the oncology unit at FHCSH (Table 3).

Factors associated with breakthrough cancer pain
In the bivariate logistic regression analysis, cancer types, stage of cancer, metastasis, the intensity of background pain, the analgesic used for background pain, and uncontrolled background pain were found to be significantly associated with BTCP (P < 0.02) and fitted for the final analysis model.The multivariate logistic regression analysis showed that cancer types (colorectal and lung cancer), metastasis, intensity of background pain, the analgesic drugs used for background pain, and uncontrolled background pain were statistically associated with BTCP (P < 0.05) (Table 4).

Discussion
Breakthrough pain is frequent in cancer patients and is often characterized by a rapid onset, short duration, and severe intensity (5,40).In this study, the prevalence of breakthrough cancer pain was 41.5% and it was consistent with previous studies that reported 39.3% (41), 41% (42), and 40.3% (43).However, it was considerably higher compared to other studies that reported 21% (44) and 29% (45).This discrepancy could be explained by the differences in the study settings.Access to quality cancer pain treatment and palliative care can be restricted in low-income countries (46).Furthermore, conflict affected areas are expected to suffer from disrupted cancer care due to damaged infrastructures, shattered supply chains, financial burdens, and evacuation of the health workforce.Similar to Northern Ethiopia, this finding is evidenced by recent studies in Ukraine and the Middle East (47,48).In contrast, the prevalence of BTCP in the current study was relatively lower than that of other studies at 64.8% (49), 69.8% (22), 73%, and 79% (50).
Among cancer types, colorectal and lung cancer were found to be significantly associated with BTCP.This finding was supported by previous studies and justified by worsening pain and discomfort during coughing, laughing, deep breathing, straining, and defecation (28,51,52).
Patients who had metastasis were highly likely to have BTCP and previous studies have supported this finding (10,49).This condition might be explained by metastatic cancer affecting multiple organs and systems.The size and rate of growth are two important factors that determine pain due to metastatic cancer.Patients could be affected by BTCP and its negative consequences, such as reduced daily activity, worsened quality of life, and dissatisfaction.Patients with metastasis are more likely to receive chemotherapy and radiation therapy, which can result in pain flareups (23).
Background pain of any intensity (mild, moderate, or severe) was associated with BTCP.It increased the likelihood of having BTCP by seven times.The association between background pain and BTCP has been reported in multiple previous studies (23,27,53,54).Cancer patients frequently experience pain, which has a favorable correlation with breakthrough pain.Untreated b a c k g r o u n d p a in c a u s e s a r e l e a se of a n e x c i t a t o r y neurotransmitter such as glutamate, which facilitates the development of BTCP (55).
If patients did not take analgesics to treat background pain, they were highly likely to develop BTCP.A large international survey in twenty-four countries showed that BTCP predominantly occurred among patients who did not take analgesics to treat background pain compared to those who did (49).Multiple previous studies also supported this finding (22, 56-58).Tumor compression against bones, nerves, and other tissues causes the release of chemicals that initiate and mediate pain.The administration of analgesics for cancer patients helps to heal the damaged tissue and prevents the  release of those chemicals (5,23).Patients who had uncontrolled background pain were more likely to complain of BTCP (59)(60)(61).This multi-center study was conducted in a research-scarce part of the world on breakthrough cancer pain.However, it did not establish temporal and causal relationships between factors and breakthrough cancer pain due to the cross-sectional nature of the study design.Statistical analysis was performed with logistic regression which could overestimate the odds ratio as the prevalence of BTCP was higher than 10%.In addition, we measured only spontaneous breakthrough pain but not incident pain during certain triggering activities such as motion.Furthermore, the inclusion of heterogeneous patients with different typologies of cancer might have limited the generalizability of the findings.

Conclusion
There is a high prevalence of breakthrough cancer pain in oncology units in Northern Ethiopia.Colorectal and lung cancer, metastasis, intensity of background pain, not using analgesics for background pain, and uncontrolled background pain were found to be significantly associated with breakthrough cancer pain.It is advised to regularly assess cancer patients for breakthrough cancer pain and provide adequate pain management.

Data availability statement
Data and materials used in this study are available and can be presented by the corresponding author upon reasonable request.

TABLE 1
Socio-demographic and clinical characteristics, and cross-tabulations with breakthrough cancer pain, N = 424.

TABLE 2
Characteristics of breakthrough cancer pain, N = 424.

TABLE 4
Bivariable and multivariable logistic regression analysis: factors associated with breakthrough cancer pain, N = 424.
COR, Crude Odds Ratio; AOR, Adjusted Odds Ratio; CI, Confidence Interval.Bold = statistically significant in the multivariate logistic regression analysis.