To investigate whether the combination of pneumatic pump with manual lymph drainage (MLD) compared to MLD alone in standard complex decongestive physiotherapy (CDP) treatment can improve outcomes in women with stage 2 lymphedema after treatment for breast cancer

Participants were randomly enrolled into 60-minute MLD or 30-minute MLD plus 30-minute intermittent pneumatic compression (IPC) followed by standardized components of CDP (compression bandaging, physical exercise, and skin care 10 times in a two-week period). Patient assessment was completed at baseline, beginning of therapy, end of therapy, and one and two months after the start of the therapy.

• The study sample (N = 27) was comprised of female patients with breast cancer who were assigned to the CDP group (n = 13) or the CDP and IPC group (n = 14).
• Mean age was 54.83 years in the CDP group and 56.60 years in the CDP and IPC group.
• Patients were included in the study if they were more than 12 months after surgery or adjuvant treatment.
• Patients were excluded from the study if they had any sign of local recurrence or distant metastases or if they were within the obligatory treatment-free period of one year.

The study took place at an outpatient setting at the Lymphedema Care Unit of the University of Szeged, Hungary.

The study used a randomized controlled trial design.

• Limb volume was measured using tape at every four centimeters.
• The percentage reduction in total arm volume at each point was calculated with the formula ∆V% = [(pre-treatment arm volume – post-treatment arm volume)/pre-treatment arm volume] × 100.
• Patients completed subjective symptom questionnaires measuring function, heaviness, tension, and pain as well as their effects on related activities.
• Percentage improvement in the subjective complains was calculated with the formula ∆S% = [(total sum of pre-treatment scores – total sum of post-treatment scores)]/total sum of pre-treatment scores] × 100.

Mean reductions in limb volumes for each group at the end of therapy and at one and two months were 7.93% and 3.06%, 9.02% and 2.9%, and 9.62% and 3.6%, respectively (p < 0.05 from baseline for each group and also between groups at each measurement). A significant decrease in the subjective symptom survey was found for both groups compared to baseline, but no significant difference between the groups was found at any time point.

The application of IPC with MLD provided a synergistic enhancement of the effect of CDP in arm volume reduction.

• The sample size was small, with less than 30 participants.
• The study took place at a single site.

More studies involved in multisite settings with large sample sizes are needed to duplicate the findings from this study.

To test the hypothesis that patients who received standardized training about pain would have fewer barriers to treatment, report lower pain, and be more likely to take prescribed opioids than patients who received a control training unrelated to pain

Patients were stratified by institution and ethnicity and then randomly assigned to control and treatment groups. Pain training included a videotape and review, with a nurse, of a handbook. The nurse focused on the concerns of the individual patient. Patients completed a checklist of eight common barriers to pain control, indicating those that he or she had wondered about. The nurse helped the patient  apply information to the patient's own situation. To assist in communication about pain, the patient then completed a checklist of things to tell the doctor. Patients were encouraged to use the checklist in communicating with the doctor and to complete a checklist any time pain was not well controlled or when other symptoms occurred. Control group patients received interventions similar in format, but these focused on nutrition. After 72 hours patients in both groups received telephone contact that reinforced training. By telephone investigators gathered patients' ratings of pain, nausea, and eating symptoms. Patients' initial training took 30–45 minutes. Follow-up calls lasted about 10 minutes. Patients completed outcome assessments at baseline and at one, three, and six months. At each time point, physicians and nurses rated patients' usual and worst pain, using the same 0–10 scale that patients used.

• Eighty-eight patients completed the study.
• Mean patient age was 55.5 years. Age range was 27–80 years.
• In the intevention group, 58% of patients were female and 42% were male. In the control group, 71% were female and 29% were male.
• The sample included multiple cancer diagnoses. Breast cancer was the most frequent diagnosis: Of all patients, 23%–31% had breast cancer. In the sample 77% had metastatic or systemic disease, and most had advanced-stage disease with a life expectancy of at least six months at the time of study entry. Of all patients, 90.5% were Caucasian. More than 50% of patients had attended some college or had earned a college degree. Baseline Eastern Cooperative Oncology Group performance status was similar in both groups, at a mean of 1.03–1.30.

• Multisite
• Outpatient
• Six urban and rural oncology clinics in the United States

Randomized blind controlled trial

• Barriers Questionnaire (BQ)
• Brief Pain Inventory (BPI)
• Memorial Symptom Assessment Scale (MSAS)
• Functional Assessment of Cancer Therapy-General (FACT-G)
• ECOG Performance Status (total daily morphine equivalent doses for pain control calculated from tables in Cancer Pain Guidelines)

From baseline the group that received training about pain had a greater decrease in barrier beliefs about cancer pain (p < 0.001), with a 0.3 point decrease, and a greater average decrease in usual pain (0.8 decrease, p = 0.03). At one month, the pain-training group reported 25% lower mean pain ratings. The pain-training group used higher doses of opioids than did the control group (p = 0.001). There were no significant differences between groups regarding worst pain, pain interference with function, or overall symptom severity. Authors noted differences between groups at one month, but there were no differences at three and six months. The pain-training group had greater narrowing of the gap between the patient’s self-assessment of pain and the nurse’s pain ratings (p = 0.001). The institution providing care was significantly associated with change in barriers to pain treatment (p < 0.001) and change in worst pain (p < 0.001). From baseline, pain interference with function (p = 0.02) was higher for nonwhites. Researchers found that, frequently, doses of medications were not re-evaluated and that patients had not told physicians about side effects or that they had stopped taking analgesics. Chart reviews presented no evidence of patients using the checklists in clinic visits or in communication with nurses and physicians.

The training intervention appeared to have only a short-term positive effect on patients' management of pain. Observations regarding related communications between patients and their professional care teams, along with significant differences in outcome by institution, point to the impact of providers in effective pain management.

• The study had a small sample size, with fewer than 100 patients.
• The sample included few nonwhite participants, and findings showed significant differences regarding ethnic covariates. These facts suggest that findings cannot be generalized to various ethnic groups.
• Although the study design included blinding of providers and those who obtained study data, authors suggest that contamination across the two groups was likely.
• Authors did not discuss patient supports or caregiver involvement in pain management.
• Authors did not describe the characteristics associated with the significant institution-related differences in outcomes.

Findings suggest that a training component may help patients self-manage pain and communicate with care providers. However, the single intervention in this study had only short-term effects that varied by institution. This points to the importance of the engagement of care providers in ongoing pain management and the need for future research to study combined patient-provider efforts. Consistent reinforcement of patient education and use of communication aids may improve long-term outcomes.

To evaluate the effects of a supervised outpatient physical therapy strengthening and conditioning program on symptoms and quality of life

The program included aerobic exercise and strength training. The first two sessions were individualized under the supervision of a physical therapist. Patients who needed additional assistance walking, transferring, or using equipment continued to attend one-hour individual sessions. Patients were placed into 90-minute group sessions. The program consisted of intensive work for eight weeks, then six months of maintenance training was offered for those who were interested. Study assessments were done at baseline and after eight weeks.

• N = 75
• MEAN AGE = 62.6 years (SD = 11.9 years)
• MALES: 34.7%, FEMALES: 65.3%
• KEY DISEASE CHARACTERISTICS: Various tumor types (breast and lung were most prevalent)
• OTHER KEY SAMPLE CHARACTERISTICS: 34.7% metastatic disease; 15.3% in active treatment; 26.7% working full- or part-time

• SITE: Single-site  
• SETTING TYPE: Outpatient  
• LOCATION: Minnesota, United States

• PHASE OF CARE: Multiple phases of care

Quasi-experimental

• Short Form 36 (SF-36)
• Six-Minute Walk Test (6MWT)
• MD Anderson Symptom Inventory (MDASI)

The results of the 6MWt were better on average at the end of eight weeks (p < 0.0001). Physical component scores on the SF-36 improved (p < 0.001) as did mental component scores after eight weeks (p < 0.005). At the end of six months, only physical component scores remained higher than reported at baseline. Fatigue (p = 0.003) and dyspnea (p = 0.007) were improved at eight weeks. After six months, fatigue (p = 0.0077), shortness of breath (p = 0.0005), and disturbed sleep (p = 0.045) were improved from baseline. Patients still in active treatment showed significantly less improvement. Those who showed the worst performance at baseline showed the greatest improvement.

The eight-week strengthening and conditioning program improved physical function, fatigue, dyspnea, and sleep disturbance in this study. Improvement was greatest among those who had the worst symptoms and physical performance statuses at baseline and among those who were not in active treatment.

• Small sample (< 100)
• Risk of bias (no control group)
• Risk of bias (no blinding)
• Risk of bias (no random assignment)
• Subject withdrawals ≥ 10%  
• Other limitations/explanation: Greater than 30% of participants withdrew from the study. There were significant differences for individuals still in treatment; however, the sample was too small to adequately analyze study sample subgroups. No information was provided about other relevant interventions that may have occurred during the six months of follow-up.

The findings of this study demonstrated the effectiveness of an exercise program on symptoms of fatigue, shortness of breath, and sleep disturbance among a variety of patients with cancer. These findings add to the large body of evidence about the efficacy of exercise.

The objective of this study was to assess the effect of optimizing hub disinfection using alcohol-impregnated port protectors by measuring the rate of central line-associated bloodstream infections (CLABSIs) and contaminated blood cultures (CBCs) in adult oncologic patient population.

The intervention involved switching traditional catheter hub care using alcohol wipes to care using 70% isopropyl alcohol-impregnated port protectors and needless neutral-pressure connectors. The intervention period results were compared to those from historical controls. Data were collected for all types of central lines, including peripherally inserted tunneled catheters and implanted ports. Line insertion techniques followed best practices for draping, skin prep, etc. Port protectors are luer lock-style caps with provide cleaning as they are twisted on and off catheter hubs. A new protector is used each time the port is accessed. Compliance was assessed by weekly observations defined as percentage of patients with catheter protectors in place.

• The sample size was 1,272.
• The age range of participants was 19–92.
• Women made up 51% of the sample; men made up 49%.
• Key disease characteristics were blood and solid cancers.

A single-site inpatient setting in West Virginia

Active antitumor treatment

Observational trial with comparison to historical controls

Observational comparison. Researchers compared CLABSI and CBC rates before and after intervention.

There was a statistically significant decrease in CLABSI rates with the use of alcohol-impregnated port protectors  from 2.3 per 1,000 central line days to 0.3 per 1,000 central line days during the intervention period (RR = 0.14, 95% CI [0.02, 1.07], p = 0.03). The rate of contaminated blood cultures decreased from 2.5% to 0.2% (RR = 0.09, 95% CI [0.01, 0.65], p = 0.002).

The use of alcohol-impregnated port protectors may help to reduce CLABSI rates. Further evidence is needed to make strong conclusions.

• Baseline sample/group differences of import   
• Risk of bias (no control group) 
• Risk of bias (no blinding)
• Risk of bias (no random assignment)
• Risk of bias(sample characteristics)
• Unintended interventions or applicable interventions not described that would influence results
• There is too much unexplained variability in this study to draw firm conclusions– types of disease and associated risk of infections are not described. 
• Varied types of catheters are included and considered together, although these are associated themselves with different risk of infection. 
• There is no information about total catheter dwell time, and whether these differed. 
• Although specific CLABSI rates may have differed, there is no information or analysis as to whether the rate of catheter removal and change differed. 
• Method of determination of compliance with specifics of use and hub changes is questionable–the fact that a hub is observed to be in place in a single point observation does not say anything about whether or not that is being changed as required for each hub access.

Suboptimal hub disinfection has a great effect on CLABSI rates. The use of alcohol-impregnated port protectors could help prevent CLABSI by eliminating this factor from the equation; however, effect sizes shown here are small. There are numerous limits to the evidence provided here which reduces the strength of the evidence and any conclusions that can be drawn; however, findings are promising. Additional well-designed studies are warranted. It would be useful to have associated cost benefit analysis, and studies comparing this approach to others to determine most cost-effective approaches to reduce CLABSI risk and rates.

The purpose of the study was to evaluate the safety and efficacy of a new rG-CSF in patients receiving chemotherapy for breast cancer.
 

Patients receiving either doxorubicin and docetaxel or docetaxel alone were entered into the study.  Patients were given rG-CSF  5 mcg/kg per day by subcutaneous injection starting on day 2 of each chemotherapy cycle, 24 hours after chemotherapy completion, that continued for either five days or until absolute neutrophil count (ANC) was greater than 1.5 x 10⁹/L. The study duration was 13 months. Severity and incidence of adverse events and antibody formation to the study drug were done. Study endpoints were incidence and duration of febrile neutropenia, duration of fever, chemotherapy cycle delays or dosage reductions, and incidence of antibiotic therapy.

• 50 patients were studied.
• Mean age of the participants was 53.54 years (SD = 10.47).
• All participants were female.
• All had breast cancer; 44% were chemotherapy naïve.

Multiple site in Lithuania.

Active antitumor treatment

Open label phase IV

• Febrile neutropenia defined as axillary temperature greater than 38.5ºC and ANC less than 0.5 x 10⁹/L.
• Common Toxicity Criteria
   

273 cycles of chemotherapy were examined. Mean duration of rG-CSF administration per cycle was 6.3 days. Eight patients withdrew from the study for various reasons. Most adverse events were associated with the chemotherapy. The most frequent grade 3–4 toxicity was neutropenia. Incidence of  grade 4 neutropenia was 47% in cycle 1 and 42% overall in patients receiving docetaxel/doxorubicin and 29% in cycle 1 and 21% overall in patients receiving docetaxel only. Most frequent study drug-related adverse events were bone pain and leukocytosis (21%), headache and musculoskeletal pain (14%), and back pain (7%).  Only bone pain was seen to be of more than mild-to-moderate severity. No neutralizing antibodies were found. Total incidence of febrile neutropenia (FN) was 14%. Mean duration of FN was 2–2.3 days. Mean duration of fever was 2.1–3.6 days depending on chemotherapy group. There was an overall incidence of chemotherapy delays or dosage changes of 1%. Overall, 20% of patients received IV antibiotics.

Overall, the study drug showed similar efficacy to other colony-stimulating factors and appeared to be well tolerated.

• Small sample (less than 100 participants)
• Risk of bias (no control group) 
• Risk of bias (no blinding)  
• Risk of bias (no random assignment)
• Findings were not generalizable
• It is not clear if analysis accounted for study withdrawals as percentages shown assume initial study sample size. Intent to treat analysis is not clearly stated. Open label design and no direct comparison with other formulations. These findings may not be generalizable to patients receiving other chemotherapeutic agents.

 Findings demonstrate effects of another G-CSF formulation.

To investigate if oral cryotherapy during myeloablative therapy may influence frequency and severity of mucositis, nutritional status, and infection rate after bone marrow transplant

Patients were randomly assigned to the cryotherapy treatment group or the usual care control group. A stratified randomization technique was used in regard to the type of transplant. Patients in the cryotherapy treatment group were instructed to suck on ice chips or rinse with ice-cold water during chemotherapy administration. The control group followed usual care without cryotherapy.

• The study reported on 78 patients aged 18 years or older.
• The mean age of the cryotherapy treatment group was 49.8 years (SD = 14.4 years), and the mean age of the usual care control group was 54.3 years (SD = 11.0 years).
• The sample was 58% male and 42% female.
• All patients had been diagnosed with hematologic or oncologic malignances. A total of 11 different diagnoses were represented.
• No significant differences existed between the cryotherapy treatment group and the usual care control group regarding age, gender, or tobacco use; the types of conditioning regimens; or the use of total body irradiation.

The study was conducted at a single-site, inpatient setting in Uppsala, Sweden.

• Patients were undergoing the active treatment phase of care.
• The study has clinical applicability for palliative care.

This was a randomized controlled trial.

• Investigators used an Oral Mucositis Assessment Score (OMAS), which was converted into the World Health Organization (WHO) scale.
• Infection rates were assessed based on neutropenic fever and use of IV antibiotics.
• Nutrition was assessed based on rate of total parenteral nutrition (TPN) and serum albumin.

• Fewer patients in the cryotherapy treatment group experienced grade 3–4 oral mucositis (OM) than in the control group (p < 0.05). However, no statistical difference was found in the severity of mucositis in the subgroup of patients receiving unrelated donor bone marrow transplant.
• No significant differences were found in weight loss between the cryotherapy treatment group and the control group.
• Fewer patients in the cryotherapy treatment group needed TPN and the number of days of TPN were fewer than in the control group. However, these differences were not statistically significant.
• The cryotherapy treatment group maintained better serum albumin levels on days 1–6 (p < 0.01) and days 7–13 (p < 0.009). 
• The cryotherapy treatment group had significantly fewer days of hospitalization (p < 0.05) in the subgroup of patients receiving unrelated donor bone marrow transplant.
• No statistical significance was found between groups related to the number of days with fever, number of positive blood cultures, or use of IV antibiotics.

Oral cryotherapy may be helpful in reducing the severity of mucositis, particularly in patients receiving autologous stem cell transplant (ASCT). Decreasing the severity of mucositis may lead to decreases in the need for TPN and better maintenance of serum albumin levels. Limited statistically significant findings were found in this study; however, it supports positive trends that favor cryotherapy use. Larger, prospective trials need to be completed.

• The sample size was small with fewer than 100 patients.
• The investigators did not clearly describe how the OMAS scale translated into the WHO scale.
• The compliance rate for using the oral cryotherapy for the unrelated  bone marrow transplant group was only three out of eight patients or less than half of the time.

Mucositis carries a high symptom burden for patients undergoing stem cell transplant. Cryotherapy may be one way to curb the effects of oral mucositis. However, this study provided no evidence to suggest that cryotherapy is the definitive way to prevent mucositis or to lessen the intensity of mucositis for all patients across the board.

Patients sucked on ice chips or rinsed with ice cold water during administration of chemotherapy. Treatment started in direct connection with and lasted until the end of the chemotherapy session.

Patients were randomized to oral cryotherapy or standard oral care. Stratified randomization was used with regard to type of BMT.

Eighty patients aged 18 and older scheduled for bone marrow transplantation (BMT). Two patients refused.

Two patients had testicular cancer; all others had hematologic malignancies (11 diagnoses evenly distributed).

Occurred from January 2002 to August 2004

Pain intensity was rated from 0–10.

Mucositis index scores  

Modified version of the Oral Mucositis Assessment Scale (OMAS)

Morphine equivalent of pain medication and duration of medications
 

Of the patients, 71%–100% managed to keep their oral cavity constantly cooled more than half the time, 58%–75% managed to keep their oral cavity constantly cooled all the time, and 7 (18%) found oral cryotherapy unpleasant. Among those seven, four (10%) found oral cryotherapy very unpleasant.

Calculated power analysis was based on days of opioids.

The experimental group had significantly fewer days with IV opioids (0.77 +/– 2.3) and complete treatment response (CTR) (2.44+/– 4.6) t = –2.053; df = 76, p = 0.045. No other differences in opioid use were observed.

Autologous BMT highest mucositis was days 9–11 (days start with chemotherapy)

Allogeneic and unrelated donor transplants peak was days 16–18

Autologous BMT experimental group (n = 62) had significantly lower mucositis score on day 10 (1.6 +/–1.9 versus 4.3 +/–5.7; t = 2.1; df = 45; p = 0.042). The experimental group also had significantly fewer days (0.06 +/– 0.25 versus 1.71 +/– 3.22, p = 0.008) and lower total dose IV opioids.

The allogenic and URD BMT group (n = 16) had significantly lower mucositis scores on day 16 (3.7 +/– 1.8 versus 11.6 +/– 6.8; t = 2.9; df = 11; p = 0.021) but  no different opioid use.

Compliance with regimen (dose of cryotherapy)

Unable to blind cryotherapy; no indication if mucositis assessors were blinded.
 

To determine whether the addition of Caphosol^® mouth rinse to a standard of care that included oral cryotherapy would decrease the incidence of oral mucositis

Patients were randomized with a computer table to the experimental or the control group. All patients received oral cryotherapy (crushed ice in the mouth during treatment), but only the experimental group received Caphosol^®. Patients in the experimental group used Caphosol^® 30 mL to rinse the oral cavity four times per day starting before high-dose chemotherapy and ending on day 21. Data were collected daily from the start of chemotherapy till day 21.

• N = 40  
• MEAN AGE = 50.4 years
• MALES: 23, FEMALES: 17
• KEY DISEASE CHARACTERISTICS: Hematologic malignancies including acute myeloid leukemia, acute lymphoblastic leukemia, chronic lymphocytic leukemia, chronic myelogenous leukemia, chronic myelomonocytic leukemia, and myelodysplastic syndrome
• OTHER KEY SAMPLE CHARACTERISTICS: There were no significant differences between the conditioning regimes of the two groups.

• SITE: Single-site    
• SETTING TYPE: Inpatient    
• LOCATION: University Hospital, Sweden

• PHASE OF CARE: Active antitumor treatment

Randomized, controlled, open-label study

• World Health Organization (WHO) mucositis grading
• Pain was assessed using the Visual Analog Scale (VAS).
• Pain medications, C-reactive protein (CRP) values, lab values, number of total parenteral nutrition (TPN) days, and hospital days were obtained from medical records.

There was no statistically significant difference between the mucositis scores, oral pain, days with TPN, use of opioids, number of hospital days, or lab values.

Adding Caphosol^® to oral cryotherapy did not provide any additional effects.

• Small sample (< 100)
• Risk of bias (no control group)
• Other limitations/explanation: In the experimental group, 15% of patients discontinued the intervention.

Oral mucositis continues to be a major complication of chemotherapy, particularity high-dose chemotherapy. Additional research with larger sample sizes is suggested because of a trend of lower pain levels, mucositis scores, and use of analgesics in patients using Caphosol^®. The 21-day use of Caphosol^® and the discontinuation of Caphosol^® because of taste or nausea are other items that merit additional study.

To evaluate the effectiveness of adding aprepitant to standard antiemetic treatment in patients receiving high-dose chemotherapy prior to stem cell transplantation

• N = 96  
• MEAN AGE = 58.1 years (experimental), 56.5 years (control)
• MALES: 70%, FEMALES: 30%
• KEY DISEASE CHARACTERISTICS: Lymphoma (38 patients) and myeloma (58 patients)
• OTHER KEY SAMPLE CHARACTERISTICS: Chemotherapy regimens, high-dose melphelan, BEAM, BEAC, and BBM

• SITE: Single site
• SETTING TYPE: Inpatient    
• LOCATION: Sweden

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

This was a randomized, blinded study.

Patient diaries were used to record nausea and vomiting. An analysis was made on an intention-to-treat basis, and differences between the groups were then analyzed. Days at the hospital, weight, and use of total parenteral nutrition also were analyzed.

Thirty-eight patients (83%) in the experimental group experienced no vomiting compared to 16 patients (36%) in the control group. This finding was statistically significant. The number of vomiting episodes was also significantly lower in the experimental arm compared to the control arm up to 17 days post chemotherapy. There were no significant differences regarding days of nausea or use of antiemetic rescue medications between the two groups. There were no significant differences noted in days at the hospital, weight and use of TPN between the two groups.

The results also showed that there was a significant difference between patients who expected nausea and those who did not. Patients who did not expect nausea had lower rates of vomiting as well as fewer days of nausea.

The addition of aprepitant to the antiemetic regimen in this patient population was well-tolerated and demonstrated a statistically significant reduction in the rate of delayed vomiting.

• Small sample (< 100)

Delayed CINV is an issue of critical importance for this patient population, and ongoing research to identify and improve symptom control and quality of life is necessary. These findings provide a springboard to conduct additional research with a larger sample size to confirm the positive impact of aprepitant on delayed vomiting.

To determine whether polaprezinc suspension in sodium alginate (P-AG) reduces the irradiation period and time to discharge after completion of radiotherapy in patients with head and neck cancer and improves the overall survival in patients with head and neck cancer who received radiotherapy

• 104 patients who accomplished 70 Gy of irradiation
• 79 patients received P-AG for prevention of oral mucositis (OM) from May 2009 to December 2014, and 25 patients received azulene gargle for prophylaxis from January 2007 to April 2009.
• Physicians checked the incidence and maximal severity of OM associated with radiotherapy every three days using a fiber-optic camera. 
• Pharmacists and nurses monitored OM and oral pain daily and recorded the results on electronic medical charts.  
• OM was graded according to the Common Terminology Criteria for Adverse Events (CTCAE), version 3.0. Pain was evaluated by a numeric rating scale. 
• The overall survival was plotted by the Kaplan–Meier method and compared statistically between the control group and the P-AG group by the Mantel–Cox log-rank test.

• N = 25 (control group), 79 (P-AG group)    
• AGE = 64.5 (49.8–75.6) (control group) and 65.7 (50.8–78) (P-AG group) 
• MALES: 90, FEMALES: 14
• CURRENT TREATMENT: Combination radiation and chemotherapy
• KEY DISEASE CHARACTERISTICS: The control group primary site of cancer was the larynx (36%), followed by the oropharynx/ hypopharynx and epipharynx. The P-AG group primary site was the larynx (43%) and oropharynx/ hypopharynx (43%).
• OTHER KEY SAMPLE CHARACTERISTICS: Of the patients in both groups, 40%–45% had no lymph node metastasis. The most common chemotherapy regimen was docetaxel (36%), followed by cisplatin 5-fluorouracil (CDDP/5-FU) and carboplatin in the control group, whereas CDDP/5-FU (34%) followed by carboplatin and docetaxel was provided for the P-AG group.

• SITE: Single site   
• SETTING TYPE: Inpatient    
• LOCATION: Gifu University Hospital, Japan

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

• Retrospective study to assess the incidence and severity of OM, the irradiation period, and the time to discharge in patients who received radiotherapy for head and neck cancer

• Common Terminology Criteria for Adverse Events (CTCAE), version 3.0, as follows: grade 1 = erythema of the mucosa in oral cavity; grade 2 = patchy ulcerations or pseudomembranes in the oral cavity; grade 3 = confluent ulcerations or pseudomembranes, or bleeding with minor trauma in the oral cavity; and grade 4 = life-threatening consequences in the oral cavity
• Pain was evaluated by a numeric rating scale.
• The Mantel–Cox log-rank test for the overall survival was plotted by the Kaplan–Meier method.

The incidence of grade 3 OM was significantly lower in the P-AG group than in the control group (16.5% versus 52%, p = 0.0003). P-AG also significantly reduced the median duration of radiotherapy (HR = 0.557, 95% CI [0.357, 0.871], p = 0.0149) and median time to discharge after completion of radiotherapy (HR = 0.604, 95% CI [0.386, 0.946], p = 0.028).

P-AG was useful for preventing OM and reducing the irradiation period and median time to discharge after the completion of radiotherapy.

• Risk of bias (no control group)
• Risk of bias (no blinding)
• Risk of bias (no random assignment) 
• Retrospective study to assess patients who received P-AG (n = 79) for prevention of OM or azulene gargle (n = 25)

Randomized and multi-institutional designs are needed to clarify the beneficial effect of P-AG on the incidence of OM and hospitalization.