STUDY PURPOSE: To evaluate the effects of massage and aromatherapy massage on symptoms in people with cancer

TYPE OF STUDY: Meta-analysis and systematic review

• FINAL NUMBER STUDIES INCLUDED = 19, 5 in meta-analysis
• TOTAL PATIENTS INCLUDED IN REVIEW = 1,274
• SAMPLE RANGE ACROSS STUDIES: 20–144
• KEY SAMPLE CHARACTERISTICS: Varied tumor types, varied settings and phases of care

PHASE OF CARE: Multiple phases of care
 
APPLICATIONS: Palliative care

Results of the meta-analysis showed no significant differences between massage and no massage for pain, depression, fatigue, or nausea. A meta-analysis of three studies showed significantly lower state anxiety with massage (MD = 18.6, p = 0.0003); however, all studies had very small samples and a high risk of bias. At longer follow-up, no differences were reported between groups in anxiety.

Insufficient evidence exists to show a benefit of massage therapy with or without aromatherapy for the relief of multiple symptoms in people with cancer.

• Limited number of studies included
• Mostly low quality/high risk of bias studies
• Low sample sizes

Evidence regarding the effects of massage and aromatherapy massage on various patient symptoms did not show substantial clinical benefit, and evidence is insufficient. However, massage is generally a very low-risk intervention that may provide very short-term benefits for some patients. Additional well-designed research is needed to determine the role of this type of intervention as part of symptom management among patients with cancer at various times in the cancer trajectory.

To evaluate the safety and efficacy of sublingual fentanyl orally disintegrating tablets for breakthrough cancer-related pain

The trial had several phases. In the initial observation phase over four consecutive days, pain intensity and pain relief were evaluated. Patients who showed an analgesic effect for opioid for breakthrough episodes were entered into the titration phase. In the titration phase (no more than 21 days long), patients received 100 mcg of sublingual fentanyl for the first breakthrough episode. The dose was increased by 100 mcg until the optimal dose was determined. Effectiveness was defined as no needed additional dose, and pain scores at 30 and 60 minutes were reduced by at least 18 mm or a third of pain levels prior to administration. Patients for whom optimal dose was determined were entered into the 21-day, double-blinded treatment phase in which three random breakthrough episodes were treated with a placebo for comparison. This was followed by a 12-week extended treatment phase. All pain scores were recorded by subjects in a diary.

• N = 37 (double-blinded phase), 13 (completed extended treatment )
• MEAN AGE = 66 years (SD = 9.2 years)
• MALES: 59.5%, FEMALES: 40.5%
• KEY DISEASE CHARACTERISTICS: Multiple tumor types; lung and colon most prevalent
• OTHER KEY SAMPLE CHARACTERISTICS: The median Visual Analog Scale pain score at baseline was 72.9.

• SITE: Multi-site  
• SETTING TYPE: Multiple settings  
• LOCATION: Japan

• PHASE OF CARE: Late effects and survivorship
• APPLICATIONS: Palliative care 

Open-label titration, double-blinded, placebo-controlled, comparative phase trial with a 12-week extended treatment phase

• 100 mm Visual Analog Scale (VAS) for pain severity (pain measured at 15, 30, and 60 minutes during breakthrough episodes)
• Patient global evaluation of relief and satisfaction on five-point scale

Overall, 90.5% of participants were titrated to an effective dose. Pain at 30 and 60 minutes during breakthrough episodes was significantly better with the sublingual fentanyl oral dissolving tablets compared to a placebo (p < 0.0001). No subjects discontinued the trial because of adverse drug effects. In the extended treatment phase, 50% of participants had dose reductions because of somnolence. Opioid analgesics were changed in 68.8% of patients because of uncontrolled pain or adverse events. The most common titrated dose was 300 mcg. The number of days needed to obtain an effective titrated dose ranged from 1–21 days with a median of three days.

Sublingual oral fentanyl dissolving tablets were effective in reducing breakthrough pain compared to a placebo.

• Small sample (< 100)
• Unintended interventions or applicable interventions not described that would influence results
• Measurement/methods not well described
• Subject withdrawals ≥ 10%  
• Other limitations/explanation: Neither the timing of pain measures nor how multiple measurements were handled for the analysis were described. There was no clear description of changes in baseline analgesia or how this was handled in the analysis.

Transmucosal opioids such as sublingual fentanyl oral dissolving tablets were effective in treating episodes of breakthrough pain. In this study, patients received this medication for extended periods, demonstrating few side effects over time.

To test the efficacy and safety of oral fentanyl in doses estimated from basal oral morphine for breakthrough pain

There was an initial observation phase in which breakthrough pain episodes were evaluated. Patients who showed effects from oral morphine entered the investigation phase. The investigation phase included three periods in which two different doses of fentanyl orally disintegrating tablets (FTs) and active controls were investigated in a random order. The dose of the FTs was determined by a conversion ratio of 1:25 for low-dose morphine and 1:50 for high-dose morphine in the observation phase. There were three episodes of breakthrough pain in each period of the investigation. FTs were given twice, and the placebo was given once. Oral morphine was used in the observation period and was given as an active control. Some studies have shown that effective doses of FTs for breakthrough pain did not correlate to round-the-clock opioid doses.

• N = 47  
• MEAN AGE = 59.1 years (range = 37–80 years)
• MALES: 60.8%, FEMALES: 39.2%
• KEY DISEASE CHARACTERISTICS: Various tumor types with lung cancer most common
• OTHER KEY SAMPLE CHARACTERISTICS: Patients were receiving daily opioid analgesics and oral morphine to treat breakthrough pain at least once every two days.

• SITE: Single site  
• SETTING TYPE: Not specified    
• LOCATION: Japan

• PHASE OF CARE: Late effects and survivorship
• APPLICATIONS: Palliative care 

Randomized, crossover, double-blinded, placebo-controlled trial with a nonblinded active drug

• Visual Analog Scale (VAS) for pain at 30 and 60 minutes during breakthrough episodes

FTs were better than a placebo for all patients (p < 0.02) and better than the oral morphine control (p < 0.001) at 30 minutes. At 60 minutes, FTs also were better than a placebo (p < 0.001). Nine subjects had adverse reactions. The most common reactions were constipation, nausea, and somnolence. One had nausea and vomiting, and one experienced respiratory depression. A temporary suspension of the investigation drug was initiated in both of these cases.

The findings of this study demonstrated that the method of dose determination used here was safe in the majority of cases, more effective than a placebo, and as effective as oral morphine for breakthrough pain.

• Small sample (< 100)
• Subject withdrawals ≥ 10%

The dose of FTs for breakthrough pain is usually determined by titrating effective doses over a period of time. During titration, the patient may have less than optimal pain control. This study provided a method of dose determination that may be useful and more timely.

Database searched was MEDLINE (1966–2001). Additional papers were found from reference lists.

Studies were included in the review if they were aimed at prevention, palliation, or reduction of radiation-induced oral mucositis in patients with head and neck cancer.

Studies were excluded if they were not in English.

More than 50 studies were included. Most were randomized, controlled trials; some were pilot or descriptive studies. 

Sample sizes ranged from 10 to more than 200.

Based on the findings of studies conducted to date, no conclusions regarding the agents and their ability to decrease the severity of radiation-induced oral mucositis were possible. Results were inconsistent. The most effective measure to treat radiation-induced mucositis was frequent oral rinsing with a bland mouthwash such as saline or sodium bicarbonate. Consistent oral care, dental care, oral assessment, and standardized oral hygiene were the suggested approaches to managing oral mucositis. Sodium bicarbonate reduces the acidity of the oral fluids immediately; it also dilutes accumulating mucus and discourages yeast colonization.

Findings related to benzydamine were inconsistent. In a trial of chlorhexidine versus benzydamine, patients reported more discomfort with benzydamine and were more likely to discontinue participation in the trial. Chlorhexidine was not effective in reducing the severity of mucositis in three double-blind, placebo-controlled trials. Two trials that examined antimicrobial activity failed to show any significant effects on the suppression of any type of oral flora using chlorhexidine.

Dose variations in granulocyte macrophage colony-stimulating factor (subcutaneous) trials make it impossible to determine whether this agent has a role in the radiation setting.

Four studies investigated the effectiveness of using topical antibiotics with a more specific spectrum for gram-negative bacteria and yeast. Two placebo-controlled, randomized clinical trials, both with fewer than 100 patients, and one case-controlled study investigated the efficacy of amphotericin B (polymyxis E, tobramycin, and amphotericin B [PTA] lozenge) to reduce the severity of radiation-induced mucositis. One study examined tetracaine and antibiotics. Additional work is warranted to determine the effects of the PTA lozenge on mucositis severity, pain severity, and dysphagia. Results for the trial were promising; however, conclusions cannot be drawn because only one study examined tetracaine.

Additional investigation of immunoglobulin and povidone-iodine are recommended.

To investigate how limited hydration in the first part of retropubic radical prostatectomy (RRP) affects the volume of intra-operative total blood loss to offer a definite hydration method during RRP

The experimental group underwent a RRP performed by the same surgeon. Eight anesthesiologists took part in the study, with each anesthesiologist completing a single case. During the first two hours of the procedure, the amount of IV fluids the patient received was limited to 1,500 ml. The control group also had RRP performed by the same surgeon; however, there was no definite methods of hydration. The two groups were matched in regard to age, serum prostate-specific antigen score, specimen weight, body mass index, performance of neoadjuvant hormonal therapy, and total IV infusion volume. No patients in either group received nerve-sparing surgery. Measurements included blood loss, IV infusion volume in the first two hours of surgery, total infusion volume, and pre- and postoperative hematocrit value.

• N = 37
• AGE = 51–79 years
• MEDIAN AGE = 66.5 years
• MALES: 100%          
• KEY DISEASE CHARACTERISTICS: Both the control and experimental groups were diagnosed with prostate cancer with elevated PSA scores. Gleason’s ranged from 6–9, with 40% of patients receiving neoadjuvant hormonal therapy prior to surgery. 
• OTHER KEY SAMPLE CHARACTERISTICS: None

• SITE: Single site  
• SETTING TYPE: Inpatient  
• LOCATION: Akashi Municipal Hospital

• PHASE OF CARE: Active antitumor treatment
• APPLICATIONS: Elder care 

• Controlled trial

• Blood loss 
• IV infusion volume in the first two hours of surgery
• Total infusion volume 
• Pre- and postoperative hematocrit value

The study demonstrated that patients receiving limited hydration remained normotensive during the first two hours of their RRP (p = 0.0343). The experiment group, those receiving limited hydration, as compared to those without IV infusion limits, also demonstrated a significant difference in overall blood loss (p = -0.0301). However, the authors further concluded that patients receiving limited hydration did not experience any adverse effects from the limited hydration (i.e., dehydration). There was no evidence in the article to support this conclusion other than one blood pressure writing during the first two hours of surgery.

Use of limited hydration in RRP during the first two hours of surgery is shown to significantly reduce blood loss (p = 0.0301). However, further studies are needed to determine if limiting hydration causes any adverse effects from the effects of dehydration.

• Small sample (less than 100)
• Risk of bias (no blinding)
• Key sample group differences that could influence results
• Findings not generalizable
• Differences in physique and shape of pelvis between Japanese and American men limits generalizability to American population; different methodologies of hydration should be considered.

The fluid balance assessment was accurate. Nurses should educate patients on surgery and the amount of bleeding to expect with the role of hydration in reducing bleeding.

To compare a coaching intervention that uses a prompt sheet with usual care to improve cancer provider and survivor communication about survivor worries

The study involved a telephone coaching session using a prompt sheet and a nurse guiding a participant in prioritizing concerns to be addressed at an upcoming oncologist visit.

• The sample was comprised of 44 survivors with breast cancer (100% female).
• Mean survivor age was 44.1 years.
• Mean age at diagnosis was 38.5 years; 20% were diagnosed at stage I, 53% at stage II, and 27% at stage III.
• Mean time since diagnosis was 5.5 years.
• All were disease-free at the time of enrollment.
• All survivors had undergone chemotherapy.
• The sample was 95% white.

• Outpatient setting
• University cancer center in Indiana

• Long-term follow-up phase
• Late effects/survivorship

A randomized controlled trial design was used.

• Breast Cancer Self-Efficacy Scale: 14-item scale with a five-point Likert-type response developed for the study (Cronbach’s alpha = 0.84, content validity confirmed by 12 breast cancer survivors, 4 clinical experts, and 2 cancer research experts)
• Spielberger State-Trait Anxiety Inventory (STAI)–State subscale
• Center for Epidemiological Studies Depression Scale (CES-D)
• Concerns About Recurrence Scale (CARS)

Most concerns centered around current symptoms, long-term effects of treatment, and recurrence of cancer. The prompt sheet and coaching did not have a significant effect on depression, anxiety, or worries. Survivor questions communicated worry about symptoms and long-term side effects experienced by young breast cancer survivors.

Self-efficacy may be a significant predictor of survivors’ state of anxiety and depression.

• Study findings are limited by sample demographics (mostly white, highly educated women).
• The study had no blinding or attentional control.
• The study’s lack of direct observation of communication between the survivor and oncologist limits the understanding of how the prompt sheet changed and/or facilitated communication.

This type of intervention may help survivors to think about their concerns prior to an office visit and organize their thoughts for the visit so that priority issues can be addressed. Whether this has an impact on outcomes has not been supported. Findings suggest that self-efficacy is predictive of role concerns and emotional outcomes such as state anxiety and depression.

To compare the efficacy and safety of a single subcutaneous injection of pegylated filgrastim with daily filgrastim as a prophylaxis for neutropenia induced by commonly used chemotherapy regimens

Fifteen centers enrolled 337 chemotherapy-naïve patients with cancer with normal bone marrow function. All patients were randomized into AOB and BOA arms and received two cycles of chemotherapy. Patients received a single dose of pegylated filgrastim 100 mg/kg in cycle 1 (AOB) or cycle 2 (BOA) and daily doses of filgrastim 5 lg/kg/day in cycle 1 (BOA) or cycle 2 (AOB). In the AOB arm, 173 patients were enrolled and 155 completed the two cycles. In the BOA arm, 164 patients were enrolled and 142 completed the two cycles. Efficacy and safety parameters were recorded. The primary endpoint was the rate of protection against grade 4 neutropenia after chemotherapy (defined as the rate at which the absolute neutrophil count [ANC] remained greater than 0.5 x 10⁹/L throughout the entire cycle).

• N = 337  
• AGE = 18–70 years
• MALES: 68 (AOB), 60 (BOA); FEMALES: 101 (AOB), 97(BOA)
• KEY DISEASE CHARACTERISTICS: Primarily breast, non-small cell lung cancer, and non-Hodgkin lymphoma; study arms essentially equal, except the BOA arm did enroll two patients with head and neck cancer; stage of disease was split equally throughout both arms
• OTHER KEY SAMPLE CHARACTERISTICS: The groups were well matched with respect to demographics, history of disease, disease severity, and ANC at baseline. The median age of patients in the study was 50 years, and almost 40% of the patients were men. Breast cancer was the most common histology. About 50% of the enrolled patients previously had undergone surgery.

• SITE: Multi-site   
• SETTING TYPE: Outpatient   
• LOCATION: Republic of China

• SITE: Multi-site   
• SETTING TYPE: Outpatient   
• LOCATION: Republic of China

• Multi-center, randomized, open-label, crossover, noninferiority study

Blood samples were collected for complete blood counts with differential on days 0, 3, 5, 7, 9, 11, 13, 17, and 21 of each cycle. The primary efficacy endpoint was the rate of protection from grade 4 neutropenia (ANC remains greater than 0.5 through the entire cycle) after chemotherapy. The secondary efficacy endpoints included the rate of grade 3/4 neutropenia, time to neutrophil recovery (defined as the time from chemotherapy administration until the ANC increased to 2.0 [after the expected nadir]), incidence of febrile neutropenia (ANC less than 0.5 and axillary temperature higher than 38.1°C), incidence of antibiotic administration, and ANC profile.

Ninety-four percent of patients receiving pegylated filgrastim or filgrastim did not develop grade 4 neutropenia. In cycle 1, the rates of protection were 89.7% (pegylated filgrastim) and 89.5% (filgrastim). In cycle 2, no episodes of grade 4 neutropenia occurred. In subgroup analysis according to chemotherapy regimens, the protective rates of pegylated filgrastim did not differ significantly from the protective rates of filgrastim.

This trial provides direct evidence that pegylated filgrastim is comparable to filgrastim in efficacy and safety when administered in support for commonly used, standard-dose, mild-to-moderate myelosuppressive chemotherapy regimens. In addition, the single injection per cycle may lead to increased patient satisfaction and compliance.

• Risk of bias (no control group)
• Risk of bias (no blinding)
• Other limitations/explanation: Not clear that the sample was at risk for neutropenia

Increased use of pegylated filgrastim across solid tumor types with mild-to-moderate myelosuppressive chemotherapy regimens. The lower frequency of visits may lead to increased emergency department visits for toxicities that normally would be caught at the daily injection visits. Education will be key in this patient population. For some patients, avoidance of daily injections may be important.

STUDY PURPOSE: To identify the best triple drug antiemetic regimen for cisplatin-induced nausea and vomiting

TYPE OF STUDY: Meta-analysis and systematic review

DATABASES USED: Not stated

INCLUSION CRITERIA: All were double-blind randomized controlled trials.

EXCLUSION CRITERIA: Not cisplatin-based therapy, no triple-drug regimen used, non-English language

• FINAL NUMBER STUDIES INCLUDED = 10
• TOTAL PATIENTS INCLUDED IN REVIEW = 7,317
• SAMPLE RANGE ACROSS STUDIES: 174–2,322 patients
• KEY SAMPLE CHARACTERISTICS: All were receiving cisplatin-based chemotherapy.

PHASE OF CARE: Active antitumor treatment

Pairwise meta-analysis was done to rank treatments for effectiveness in terms of complete response (CR) rate. Ranking for best results was: (a) netupitant, palonosetron, and dexamethasone (NEPA); (b) fosaprepitant, ondansetron, and dexamethasone; (c) palonosetron and dexamethasone; (d) fosaprepitant, granisetron, and dexamethasone. However, comparisons did not reach statistical significance. NEPA also ranked highest in percentage of cases with no nausea. The regimen of aprepitant, granisetron, and dexamethasone ranked highest in the side effect of constipation. The regimen of rolapitant, ondansetron, and dexamethasone ranked highest for delayed nausea control and fewest side effects. No significant differences existed across regimens in side effects.

Findings suggest that triple drug regimens, including NEPA, may be most effective in chemotherapy-induced nausea and vomiting (CINV) prevention and the prevention of delayed nausea, though actual differences across all triple drug regimens were not statistically significant.

The authors noted that some evidence reveals that the efficacy of various triple drug regimens may depend upon the tumor type rather than the antiemetic regimen. In general, all triple drug antiemetic regimens are shown to be effective for CINV management, and variation across regimens exists regarding their efficacy for nausea and response in the delayed phase, in particular. Further research is needed to identify comparative effectiveness for various regimens with analysis by tumor type as well as type of chemotherapy. In practice, given potential differences in effect, regimens for individual patients should be planned according to individual patient responses and risks.

To compare the efficacy and safety of two different 5-HT₃ receptor antagonists

Patients were randomized to one of two groups. In the first group, patients received ramosetron for prevention of chemotherapy-induced nausea and vomiting (CINV) in cycle 1 of highly emetogenic chemotherapy (HEC) and then ondansetron in cycle 2. Patients in group 2 received ondansetron first, followed by ramosetron.

The study consisted of 50 patients.

The study was conducted in China.

This was a prospective, randomized, crossover trial.

Nausea, emesis, and loss of appetite were assessed, but the authors did not describe how this data was collected or by whom.

No significance was reported for any measure between groups at any time point.

No significant differences were reported between the two groups on any measure at any time point

• The sample size was small.
• The measures were not clear.

To evaluate whether participation in a problem-solving intervention influences level of caregiver assistance with patient symptoms, caregivers’ depressive symptoms, burden mastery, and caregiver/patient communication at 10 and 16 weeks postparticipation

Caregivers in the dyads assigned to the intervention group received three telephone calls from a master’s-prepared nurse to assist them in assessing and managing patient symptoms. Both the intervention and control groups received written materials on these topics, and control group caregivers received calls from a non-nurse coach who reminded them of applicable content sections of the written materials. Measures were obtained at baseline, 10, and 16 weeks. Both groups received a symptom management toolkit containing written materials on symptom assessment, communication, and symptom management. Intervention group caregivers received three phone calls from a nurse to assist in identifying and managing symptoms; control group caregivers received calls from a non-nurse coach who reinforced the material in the toolkit.
 

• The sample was comprised of 225 caregiver/patient dyads.
• Mean caregiver age was 53.8 years (SD = 12.7) in the intervention arm and 56.1 years (SD = 13.1) in the control arm. Ranges were not reported.
• The intervention arm had 47 males (SD = 41.96) and 65 females (SD = 58.04); the control arm had 43 males (SD = 38.05) and 47 females (SD = 41.96).
• Patients were 40 years or older, without cognitive impairments; were English-language users; had touch-tone telephone service; had stage III or IV solid tumors; were not on hospice; and had a family caregiver.
• Patients had to be actively receiving chemotherapy during the study and report both pain and fatigue within seven days prior to recruitment.
• Patients under the care of a psychiatrist or psychologist were excluded.

• Multisite 
• Home setting
• Telephone intervention

Active antitumor treatment phase

A randomized controlled trial design was used.

• Cancer Symptom Inventory
• Caregiver Symptom Involvement (developed for study)
• Center for Epidemiologic Studies–Depression Scale (CES-D)
• Caregiver Reaction Scale (caregiver burden)
• ENRICH Marital Inventory–Family Communication Subscale
• Pearlin Mastery Scale

The statistically significant effect observed in the study related to a differential effect of depression in the intervention arm at the 10-week time point, where caregivers with lower levels (less than 16 on the CES-D) were twice as likely to provide an intervention for patient symptoms than those with a higher depression score (OR = 1.99, 95% CI = 1.45–2.76). Caregiver self-esteem was also statistically significantly different in the intervention arm (p = 0.04), but the authors noted that in the clinical context, this finding was likely due to chance.

Although no significant differences were noted overall between the nurse-led intervention and the control group, knowledge was gained regarding the impact of caregiver depressive symptoms on the degree of interventions offered to the patient with cancer experiencing symptoms. Future research may focus on tailoring interventions based on dynamic characteristics such as degree of caregiver distress concurrent to increasing patient symptom needs.

The study had risk of bias because the sample was described as primarily Caucasian: The authors cited literature noting variances in caregiver emotional responses by race, which might have implications given the findings associating depressive symptoms and caregiver responsiveness.

The authors speculated that caregivers with higher levels of depressive symptoms may be less able to act on behalf of their family members who are patients. In practice, nurses should assess whether caregiver distress may impact outcomes such as medication adherence and effective symptom reporting and management to avoid impending crises.