Introduction

Cancer patients often experience multiple physical and psychological symptoms during and after an aggressive cancer treatment. Hepatocellular carcinoma (HCC) is a worldwide health problem leading to more than 250,000 deaths annually, and it is now the leading cause of cancer-related death in eastern Asia, North America, and Europe [1, 2]. Surgical intervention, resection, and transplantation are the optimal and best option for cure. However, about 60–70 % HCC patients are diagnosed at an intermediate or advanced stage [3, 4]; by then, nonsurgical treatment is the only option left. Nonsurgical treatments include systemic and infusional chemotherapy, hepatic ligation or embolization, percutaneous ethanol injection, cryotherapy, radiolabeled antibodies, and radiofrequency ablation [3–5]. For HCC patients with unresectable diseases, the goal of palliative treatment is to slow down disease progression, control symptoms, improve quality of life, and prolong survival.

Transcatheter arterial chemoembolization

The Barcelona Clinic Liver Cancer (BCLC) staging system, which has been derived from cohort studies and randomized controlled trials (RCTs), is widely recognized and endorsed [5–7]. BCLC links stage stratification with a recommended treatment strategy and defines the standard of care for each tumor stage. According to the BCLC system, transcatheter arterial chemoembolization (TACE) is recommended as the first-line palliative treatment for unresectable intermediate-stage HCC. Many studies, including a meta-analysis involving six RCTs, have demonstrated that TACE can greatly improve survival of HCC patients compared with other supportive care or suboptimal therapies [8–11]. Clinically, about 60–70 % HCC patients were diagnosed at intermediate to advanced stages and had lost the chance for operation [3, 4]. Patients with intermediated-stage HCC and partly preserved liver function, with no cancer-related symptoms, and no evidence of vascular invasion or extrahepatic spread are considered candidates for TACE. In general, TACE is not suitable for patients with decompensated liver cirrhosis (ascites and/or jaundice) due to the high risk of serious treatment-related complications.

HCC deprive approximately 80–85 % of their blood supply from the hepatic artery, whereas normal hepatic parenchyma is supplied primarily by the portal vein as well as the hepatic artery. TACE is a hepatic-directed therapy taking the advantage of relatively selective hepatic arterial tumor vascularization. The goal of TACE is to deliver a high dose of chemotherapeutic drug and embolization agent to increase local chemotherapeutic dwell time and induce tumor necrosis and control [12]. TACE is often conducted by an interventional radiologist with at least a 3-year experience in TACE. To reconfirm the site of the HCC and identify the nourishing artery, all patients are required to receive angiography of the celiac artery and superior mesenteric artery and computed tomography (CT) during hepatic arteriography and arterioportography. Prior to TACE, highly selective catheterization is performed using a 2.5-F microcatheter to achieve complete occlusion of the feeding arteries and to preserve liver function. TACE is performed by injection of a mixture of 5–10 mg mitomycin, 30–50 mg doxorubicin, and 5–10 ml of lipiodol. Embolization is subsequently done with gelfoam powder and a small amount of contrast medium under fluoroscopic guidance.

Monitoring of the tumor response is paramount to determine the time interval of repeated treatments and to identify treatment failure. HCC patients are asked to recheck the tumors by CT scans every 1.5 months after TACE for half year, every 3 months for 1 year, and every 6 months thereafter. Patients with tumor recurrence are advised to receive repeated TACE if their liver functions are good enough and there are no contraindications. The shortest time interval between the two TACE treatments is about 1.5 months so as to protect liver function.

Symptoms in HCC patients receiving TACE

HCC patients receiving TACE often suffer from various symptoms resulting from both the primary disease itself and treatments; many symptoms are common to cancer patients but some are unique. Nausea, vomiting, fatigue, and lack of appetite are the general symptoms caused by chemotherapy. HCC patients receiving TACE experience more symptoms, such as severe abdominal pain originating from swelling of liver capsule or cholecystitis, stomach discomfort due to acute gastric mucosal lesion, dyspnea, and upper gastrointestinal bleeding [13, 14]. And due to high recurrence rate of HCC, these patients often need repeated TACE for residual viable tumor or local recurrence. Although TACE is less invasive, widely indicated, and relatively safe, it may also cause some severe symptoms and complications, sometimes leading to serious clinical situations and even death. Patients who need to receive repeated TACE have to experience multiple symptoms each time, and these symptoms can impact the patients both psychologically and physiologically. If not well managed, they will reduce patient compliance to TACE.

Symptom clusters in HCC patients

Kim et al. (2005) defined a symptom cluster as a stable group of two or more concurrent symptoms related to each other and independent of other symptoms or symptom clusters [15]. Compared with a single symptom, symptom clusters have more complicated influences on patient outcomes, quality of life (QOL) and follow-up treatment. Many researchers, using various statistical analyses, had tried to clarify the concept of symptom clusters manifested during and after treatments of various cancers. However, only one study from Korea reported the symptom clusters in 180 HCC patients [16], in which the symptoms were assessed via a symptom checklist [17], and at the same time, the patients completed the hospital anxiety and depression scale and the functional assessment of cancer therapy-hepatobiliary. Factor analysis was used to identify the symptom clusters based on the severity of symptom experiences. The authors found four symptom clusters in HCC patients: pain–appetite, fatigue-related, gastrointestinal, and itching–constipation. The findings of this study suggest that various symptom clusters may contribute to the poor QOL and depressive anxious moods in HCC population. However, the sample of this study was heterogeneous in terms of different treatments. Treatment-related symptoms were not clear and the implications for symptom management were limited.

Palliative cancer care for HCC patients receiving TACE is to provide them with symptom relief and better quality of life. However, most researchers are concerned with technological innovation of TACE, such as comparing the efficacies of TACE using different anti-cancer drugs [18], evaluating whether traditional Chinese medicine can improve immune response for unresectable hepatocellular carcinoma after TACE [19], and exploring the effectiveness of transcatheter arterial chemoembolization and microsphere embolization for HCC patients [20]. Few studies have focused on symptom changes before and after TACE, and no studies have explored whether multiple concurrent symptoms of HCC patients receiving TACE can be grouped into clusters. Symptom management falls far behind the technological innovation in HCC patients receiving TACE.

Clinical management of symptoms is based on the understanding of the multiple dimensions of cancer- and treatment-related symptoms. A lack of sufficient knowledge regarding symptom prevalence, severity, and interference is a factor affecting symptom management for HCC patients receiving TACE. How to alleviate symptoms is still a core issue in nursing care for HCC patients receiving TACE. In the present study, we aimed to observe symptom experiences in HCC patients before and after TACE, identify whether multiple symptoms can be clustered into symptom groups, and explore the interference of these symptom and symptom clusters on daily life of HCC patients receiving TACE.

Methods

Sample and settings

A total of 185 HCC patients receiving TACE were invited to participate in this study. The diagnosis of HCC was made based on ultrasonography, dynamic CT, magnetic resonance imaging, and angiography and/or pathologically by biopsy specimens. Additional inclusion criteria for the patients included: (1) the Child–Pugh class was A or B; (2) HCC < 50 % of hepatic volume; (3) the patients received their first episode of TACE; and (4) adult patients (≥18 years of age) who were able to read, write, and understand Chinese and able to give written informed consent. Patients who could not understand the intent of the study, who refused to participate, or who were currently diagnosed with a major psychiatric illness were excluded. All the patients were recruited from our teaching hospital—Eastern Hepatobiliary Hospital. Twenty-nine patients withdrew due to mental or physical exhaustion and inability to understand the questionnaires; and one patient died due to upper gastrointestinal bleeding. The questionnaire return rate was 84 %.

Measures

The demographic data of subjects were collected using an interview tool, which includes information about sex, age, marital status, education level, and religious beliefs. The clinical information was collected through interviews and verified by reviewing patient medical records, including information about BCLC stage, tumor emboli, liver function, chemotherapeutics, embolic agents and the puncture site.

The liver function was assessed using the Child–Pugh status. The patients were classified as Child–Pugh class A (5–6points, mildly impaired), B (7–9 points, moderately impaired), and C (10 or more points, severely impaired).

Symptoms of patients were assessed using the M. D. Anderson Symptom Inventory (MDASI-Chinese) [21] and a symptom checklist particularly for hepatobiliary cancers; the latter includes six symptoms: abdominal distension, diarrhea, fever, itching, weight loss, and jaundice. MDASI-C is rated with an 11-point numeric rating scale (NRS), with 0 indicating “not present” and 10 indicating “as bad as you can imagine.” Six interference items were assessed using an 11-point NRS (i.e., 0 “does not interfere” to 10 “interferes completely”). Items of the symptom checklist particularly for hepatobiliary cancers are also rated with an 11-point numeric scale, from 0 (not at all) to 10 (as bad as you can imagine). We identified the initial symptom items by literature review and fifteen experienced doctors and nurses with at least 10-year clinical experience. Then, 250 HCC patients were tested in a pilot study. The validity of the symptom checklist was verified by a content validity index of 0.911, and the Cronbach's alpha was 0.933.

Procedure

Subjects who met the inclusion criteria were invited to participate in the study. At the beginning of study, we explained to the subjects the purpose of our study and obtained their verbal consent. We promised that they had the right to withdraw at any time without any penalty or adverse effect on their treatment. The questionnaires were administered as part of routine clinical monitoring and the entire process was supervised by our trained research assistants. Patients completed the questionnaires independently after explanation of items by the researchers who received a uniform training before the study. Symptoms were separately assessed at two time points—T1: before TACE (the day patients were admitted to hospital); T2: after TACE (about 48 h after TACE, because within 48 h after TACE there may be tumor necrosis and swelling of liver capsule and the patients have the highest fatigue level [22]).

Statistical analysis

Descriptive statistics was used to summarize the social, demographic, and treatment data. Principal component analysis (PCA) with varimax rotation was used to examine whether the presence of interrelationship among symptoms exist and to identify the symptom clusters. Initial analyses showed that symptoms with an extremely low prevalence rate might interfere with the analysis process (e.g., restricted variance) and interpretation of results. Furthermore, the symptoms with a low prevalence do not reflect the key symptoms in a cluster. Therefore, uncommon symptoms (prevalence <10 %) were excluded from the analyses [23]. At Time 1, before TACE, ten symptoms (nausea, vomiting, problem of memory, shortness of breath, drowsiness, numbness, diarrhea, jaundice, itching, and fever) were excluded. At Time 2, after TACE, three symptoms (problem of memory, numbness, and diarrhea) were excluded. The number of significant principal components was selected with an eigenvalue higher than 1.00. Symptoms with a salient loading on a factor were considered to be a symptom cluster. The number of clusters was determined by the Cronbach's coefficient (internal consistency) for each factor higher than 0.65, the case number of each cluster, and clinical phenomena [21]. The internal consistency and reliability of the derived clusters were assessed with Cronbach's alpha. Spearman's correlations were obtained to explore the correlation between symptoms and symptom interference.

Results

Patient characteristics

The subjects consisted of 155 HCC patients (142 males, 13 females), and 95 % of them were married. The average age of the participants was 53.25 (SD = 10.33; ranging from 22 to 77). The time since diagnosis ranged from 1 to 4 months, with a median of 2 months. Thirty-seven patients had tumor emboli. The puncture site was the femoral artery of the right leg, and the most common chemotherapy medicine and embolism agents were gelfoam, lipiodol (10-30 ml), Adriamycin (ADM,30-50 mg), and Cisplatin (CDDP,40 mg-80 mg). Details of the demographic and medical characteristics are summarized in Table 1.

Table 1 Sample demographic and clinical characteristics (n = 155)
Full size table

Prevalence and severity of symptoms

Table 2 lists the prevalence and severity of symptoms in MDAS-C and the HCC symptom checklist. Before TACE, the five most severe symptoms, ranked in order, were fatigue (3.40 ± 2.26), distress (3.35 ± 2.60), sadness (3.01 ± 2.66), sleep disturbance (2.63 ± 2.57), and lack of appetite (2.26 ± 2.38). After TACE, fatigue (4.88 ± 2.31) was the most serious symptom, followed by sleep disturbance (4.80 ± 2.25), distress (4.59 ± 2.32), sadness (4.45 ± 2.16), lack of appetite (4.25 ± 2.51).

Table 2 Prevalence and severity of symptoms in MDAS-C and HCC symptom checklist
Full size table

Symptom clusters of HCC patients undergoing TACE

Time 1: before TACE

Before TACE, only nine symptoms were included for PCA, including pain, fatigue, sleep disturbance, distress, sadness, lack of appetite, dry mouth, abdominal distention, and weight loss. As mentioned previously, to understand the latent constructs of the nine included symptoms before TACE, we conducted an exploratory factor analysis using the principal component method with varimax rotation. Two factors had eigenvalues greater than 1.00, accounting for 60.92 % of the total variance. Factor loadings of the nine symptoms with varimax rotation are shown in Table 3. Factoring analysis identified two symptom clusters: Cluster 1 psychological symptom cluster (including distress and sadness), accounting for 26.45 % of the total variance; Cluster 2 sickness symptom cluster (including pain, fatigue, sleep disturbance, lack of appetite, dry mouth, abdominal distention and weight loss), accounting for 34.37 % of the total variance. The internal reliabilities of the two clusters using Cronbach's alpha coefficient were 0.911 and 0.833, demonstrating good internal consistency (Table 3).

Table 3 Factor loadings and final communality from the PCA of the symptoms (scale 0–10) before TACE
Full size table

Time 2: after TACE

After TACE, sixteen symptoms were included for PCA and four clusters were finally identified, which accounted for 66.43 % of the total variance. Cluster 1 psychological symptom cluster was composed of distress, sadness, and sleep disturbance; Cluster 2 sickness symptom cluster consisted of fatigue, pain, lack of appetite, abdominal distention, weight loss, fever, dry mouth, drowsiness, and shortness of breath; Cluster 3 upper gastrointestinal symptom cluster included nausea and vomiting; and cluster 4 liver function impairment symptom cluster included jaundice and itching. The internal consistencies for the four clusters were 0.95, 0.79, 0.99, and 0.96, respectively, showing good internal consistency (Table 4).

Table 4 Factor loadings and final communality from the PCA of the symptoms (scale 0–10) after TACE
Full size table

Association of symptom and symptom cluster severities with their interference on daily life of HCC patients

Wilcoxon signed-ranks test was used to determine whether the differences were statistically significant for patient symptom interference before and after TACE. As shown in Table 5, the total interference with daily life rose by 0.52 point after TACE (Z = 3.64; p < 0.001). Significant differences were found in all the items of symptom interference after TACE except for the item of relations with others. The top three symptom interference items were work, enjoyment of life, and mood. Spearman correlations were performed to test the relationship between single symptom and interference. Before and after TACE, the symptoms of distress, sadness, fatigue, sleep disturbance, and lack of appetite were all significantly associated with the total interference, with the collection coefficients ranging from 0.443 to 0.615.

Table 5 Differences in symptom interference before and after TACE
Full size table

Scores for the clusters (before and after TACE) were computed by adding the scores of each symptom within each cluster. Before TACE, we found that both the two symptom clusters were significantly related to symptom interference, while the psychological symptom cluster (including symptoms of distress and sadness) was more closely correlated with total interference than sickness symptom cluster. The collection coefficients ranged from 0.638 to 0.560. After TACE, sickness symptom cluster became the most serious problem, followed by psychological symptoms cluster, upper-gastrointestinal symptom cluster, and liver function impairment symptom cluster (shown in Table 6).

Table 6 Relationships between symptom clusters and symptom interference
Full size table

Discussion

This study provides the information on symptom experiences and the impact of symptom clusters on daily life of HCC patients receiving TACE. Before TACE, the prevalence and severity of symptoms were both at lower levels, but they rose quickly after TACE, especially physical symptoms such as pain, nausea, vomiting, jaundice, and drowsiness. HCC patients often require repeated TACE for residual viable tumors or local recurrence; some patients eventually refuse TACE because they cannot tolerate the repeated painful symptoms caused by TACE. We realize that if not well managed, these symptoms can cause interruptions or cessations of cancer treatment.

Our results also showed that most HCC patients undergoing TACE were suffering from multiple symptoms which can be grouped into clusters. Across the treatment trajectory, psychological symptom cluster and sickness symptom cluster remained relatively stable, while two new symptom clusters emerged after TACE: upper gastrointestinal and liver function impairment symptom clusters. The psychological cluster was present both before (i.e., Time 1) and after TACE (i.e., Times 2), but the symptom of sleep disturbance transferred from sickness symptom cluster before TACE to psychological symptom cluster after TACE. A possible interpretation was that sleep disturbance is a physical symptom as well as a psychological symptom. The sickness symptom cluster before TACE changed a little after TACE, with fever, shortness of breath, and drowsiness entering in the cluster. The upper gastrointestinal symptom cluster and liver function impairment symptom cluster were both present after TACE. The two symptom clusters were directly related to TACE, because during TACE, most chemotherapeutic and embolization agents stay in the tumor, but some still remain in normal liver tissue around the tumor, thus damaging the liver function, and some agents enter the blood vessels of stomach, inducing nausea and vomiting.

Unrelieved symptoms can have a negative impact on the daily life, functional status, and quality of life [24]. Level of symptom interference is an important indicator not only for assessing patients' quality of life, but also for changing, delaying, or discontinuing a cancer treatment. In our study, we found that the psychological symptoms contributed more significantly to symptom interference than physical symptoms before TACE, which might be specific for HCC. Hepatocellular carcinoma has a high mortality and a quiet early clinical manifestation, and it is hard for patients to accept the fact that they are victims. The nurses should closely monitor the mental status and emotion of patients before TACE and take appropriate measures to alleviate their anxiety, fear, and sadness. After TACE, physical symptoms emerge concurrently. The interference of these physical symptoms is more prominent than before TACE. So after TACE, nurses should effectively manage the physical symptoms in addition to the psychological ones.

Taken together, we have the following suggestions on how to manage symptoms before and after TACE:

  • First, doctors and nurses should attach more importance to the comprehensive symptom assessment and management in HCC patients receiving TACE. Clinically, we often pay attention to only the symptoms such as the pain, fever, nausea, and vomiting. However, HCC patients also have some prevalent and severe symptoms such as fatigue, sleep disturbance, and lack of appetite, which we often ignore. The sickness symptom cluster in HCC patients demonstrates that when patients have pain, they also suffer from fatigue, sleep disturbance, lack of appetite, and so on. Pain can lead to sleep disorder and fatigue directly or indirectly, and pain intervention can alleviate pain–sleep disorder–fatigue symptom cluster, finally improving patient quality of life. This suggests that nurses should also consider sleep and appetite during evaluation and management.

  • Second, priority should be given to psychological care before and after TACE in HCC patients. In our research, distress and sadness were always at a high level and had great influence on patients' daily life. Nurses should explain the treatment technique to patients before TACE, teach them the methods to evaluate symptom severity after TACE, and make them realize that they are not alone.

  • Third, early therapeutic interventions should be taken to control the side effect symptoms. Our investigation showed that pain, nausea, vomiting, and jaundice were prevalent and severe after TACE; if measures are taken in advance, we believe that these suffering symptoms can be mitigated. For instance, we introduced the concept of preemptive analgesia in HCC patients and gave them some medicine to protect gastric mucosa or liver function. These measures may reduce the fear of TACE treatment and increase patient compliance to treatment and care.

  • Fourth, researches on symptom management should be strengthened. Although the clinical implications of symptom clusters are limited and very tentative, clinicians can use the concurrent tendency of symptoms in a cluster to assess symptoms and provide preparatory information on the symptom experiences of TACE. Some studies reported fatigue, insomnia, and pain as “sentinel” symptoms which are likely to have a major effect on functional status and overall symptom burden [25]. As more of these symptoms are present, negative outcomes become more likely. Proper handling of the sentinel symptoms might be the most efficient and effective way for management of specific symptoms in HCC patients.

Limitation

Unfortunately, our study has the following limitations. First, since we excluded the symptoms of low incidence (incidence < 10 %), there may be some bias, and we need larger sample study in the future. Second, differences in cancer stage, gender, and age were not considered in our study. Thus, any confounding factors cannot be extracted. Further work can be done to better understand symptom cluster occurrence and patterns. Third, we did not record any data on comorbidity or medication usage, and we may loss some valued information. Forth, our study was only a cross-sectional study. Exploring the long-term effectiveness of TACE in easing symptoms and improving the quality of life will be a problem worth studying considering the variations of symptom clusters during treatment.