The objective of the study is to compare the effects of oxygen application and use of opioid treatment on ventilation and palliation of chronic dyspnea in hypoxic and non-hypoxic palliative care patients.

Four liters per minute of oxygen were given via nasal cannula. After 60 minutes, patients rated dyspnea, and respiratory parameters were recorded for comparison to baseline. Patients initially received immediate release opioids every four hours and rescue doses as required for breakthrough dyspnea. Patients who had been pre-treated with opioids were switched to oral morphine equivalent doses, and the opioid dose was titrated to achieve a tolerable and stable level of symptoms. When stable, patients were changed to sustained release opioids.

The study reported on a sample of 46 patients. The median age was 66.5 years, with a range of 40-90 years, for hypoxic patients and a median age of 70.5 years, with a range of 40-86 years, for non-hypoxic patients. Twenty-three patients were females; 9 were hypoxic, and 14 were non-hypoxic. Twenty-three patients were males; 9 were hypoxic, and 14 were non-hypoxic. In the hypoxic group, participants were diagnosed with end-stage cancer of multiple types. Participants were included if they experienced dyspnea at rest, had a hemoglobin level of greater than or equal to 10 g/dl measured within 2 weeks, and had normal cognitive function. Of the participants included, 18 were noted to have been pre-treated with opioids for pain control, while 28 were opioid naive. In addition, four participants were noted to have a medical history of chronic obstructive lung disease. Seventeen patients reported intermittent pre-treatment with oxygen therapy (2-6 L/min).

The single-site study was conducted on a palliative care unit in Germany.

Patients were undergoing end-of-life and palliative care.

Prospective non-randomized study

• Numeric rating scale (NRS) of intensity of dyspnea (scale of 0-10)
• Karnofsky Performance Status Score to assess baseline level of function
• Transcutaneous measurement of carbon dioxide partial pressure, oxygen saturation, and pulse rate via an earlobe sensor

During 60-minute oxygen insufflation with the 4 L/min nasal cannula, no decrease in dyspnea was noted among the hypoxic and non-hypoxic patients. No significant correlation was seen between intensity of dyspnea and oxygen saturation. A significant increase was seen in SaO₂ in hypoxic patients after opioid application (P < 0.0001), and a significant decrease was seen in respiratory rate in both groups. In non-hypoxic patients, respirations in the opioid naïve ranged from 38 (SD = 5.6) per minute to 27 (SD = 4) per minute (P = <0.0001) after 120 minutes, while patients who were pre-treated with opioids ranged from 37 (SD = 4.5) per minute to 27 (SD = 3.4) per minute (P = 0.003) after 120 minutes.

• Use of opioids for the reduction of dyspnea was more effective than oxygen insufflation, even in hypoxic patients.
• Oxygen did not reduce dyspnea.
• Opioids appeared to reduce dyspnea both in opioid-naïve patients and those who had previous medication with opioids.

The study had a small sample of less than 100. Baseline measurement of dyspnea intensity, SaO₂, PaCO₂, and tcpaCo₂ during 60-minute oxygen administration obtained from the 17 participants who were known to have used as much as 6 L/min nasal cannula at home calls into question the accuracy of comparative data because study intervention used only 4 L/min nasal cannula. The results only pertain to patients experiencing “chronic” dyspnea and do not relay or compare the effects of increasing oxygen concentration or modes of oxygen delivery (i.e., greater than 4 L/min, face mask, FiO₂, etc.) in management of acute, chronic, or breakthrough dyspnea in hypoxic and non-hypoxic patients.

Oxygen therapy for the management of chronic dyspnea in patients with cancer with advanced disease may not be a cost-effective intervention and has no established long-term benefits for symptomatic relief of work of breathing. Intermittent use of opioids for the safe improvement of symptomatic dyspnea may be a better alternative with minimal likelihood of resulting respiratory depression.

To determine the effectiveness and frequency of manual lymphatic drainage (MLD) in patients with lymphedema secondary to advanced stage cancer undergoing palliative care

Patients were included in the study if they had lymphedema-related symptoms, pain, and dyspnea. The intervention strategy consisted of utilizing demographic information and disease-related data to calculate and compare the effects of MLD on lymphedema-related symptoms. MLD was performed daily on patients until discharge, and the effects were documented.

• The study sample (N = 90) was comprised of 36.7% male and 63.3% female patients.
• Mean age was 65.5 years. 
• Patients had a variety of cancer types, including thoracic, gastrointestinal, genitourinary, breast, and pancreatic. 
• Seventy-four percent of patients reported pain, 88.1% reported somatic pain, 11.9% reported somatic and neuropathic pain, and 25.6% reported dyspnea.

The study took place in an in-patient setting in a palliative care unit from January to December 2007.

Patients were undergoing active treatment for lymphedema. The study has clinical applicability for palliative care.
 

The study used a retrospective, reflexive control design.

• A numeric rating scale was used to measure pain and dyspnea.
• A four-point Likert scale assessed lymphedema reduction.
• The Karnofsky Performance Scale Index determined the level of functional impairment.
• SPSS Statistics evaluated data.

Sixty-seven patients reported pain (59 somatic, 8 somatic and neuropathic) and 23 patients reported dyspnea. Patients in the study received an average of seven MLD treatments. MLD treatment lasted an average of 41.3 minutes. The treatment was tolerated well in 92.2% of patients. Ninety-four percent of patients reported a reduction of pain and 73.9% of patients reported a reduction of dyspnea  post-treatment. Four of the eight patients with reported neuropathic pain had to discontinue the MLD intervention because of increased intensity of pain during treatment. The patient self-report of lymphedema reduction on the Likert scale was reported as 18.9% as little, 64.4% as good, and 16.7% as good.

The study suggests a positive correlation between a significant decrease in pain and dyspnea after MLD in patients with advanced disease or cancer undergoing palliative care but increased pain with MLD in patients with neuropathic pain.

• The sample size was small, with less than 100 patients.
• The study had a risk of bias because there was no control group, blinding, random assignment, or an appropriate attentional control condition. 
• Because opioid analgesics were given a few days prior to the intervention (although they were discontinued before beginning), patients' perception of pain may have been altered.

In patients undergoing palliative care, the top priority is comfort. Given the positive outcomes of the study, nurses may recommend that their patients partake in MLD as a mechanism to alleviate burden because of their disease process, if the pain is not of neuropathic origin. It should be recognized that some patients experienced increased pain with MLD, so nurses need to be aware of assessing these effects if MLD is used. Further studies should be completed using a more rigorous study design and larger sample size to further support findings reported in the study.

The objective of this study was to assess the safety of opioid and benzodiazepine combination for dyspnea management in patients receiving palliative care.

Patients in a palliative care unit were provided morphine or hydromorphone and lorazepam enterally every four hours, and rescue doses were given as needed every 15 minutes according to a titration schema. All patients were given 1 mg lorzepam sublingual with the first opioid dose during the clinical stay. Ratings of dyspnea were recorded at rest and on light exertion. PaCO ₂ and SaO₂ were monitored with earlobe sensors. Measurements were taken at baseline for 15 minutes after patients were admitted to the palliative care unit and for at least 240 minutes after starting the opioid and lorazepam combination. 

• The study reported on a sample of 26 participants.
• The mean age was 66 years (SD = 13.6 years).
• Of the sample, 53.8% were males and 46.2% were females.
• Patients had a variety of cancer types, with lung and breast the most frequent.
• All patients had advanced disease, with a mean survival time of 51.5 days.
• The mean length of stay in the palliative care unit was 12.8 days.
• No patients were on oxygen prior to the study.
• SaO₂ at study entry ranged from 85%–100%. 
• Of the participants, 42.3% were opioid naïve at study entry.
• All patients had at least moderate dyspnea, defined as at least 4 on the 11-point scale.
   

This single-site study was conducted in an inpatient setting in Germany.

• Patients were undergoing end-of-life care.
• The study has clinical applicability for end-of-life and palliative care.
   

The study was a prospective, nonrandomized trial.

• Numeric dyspnea rating scale (0–10)   
• Pulse oximetry
   

Mean morphine dose was 8.4 (SD = 7.2), and mean hydromorphone dose was 30 (SD = 35) morphine equivalents. Respiratory rate was significantly reduced 60 minutes after the combination of medications was delivered (from 40–30, p < .001), and dyspnea at rest declined from mean of 6.2 to 4.1 after 30 minutes and to 1.2 after 120 min (p < .001). No significant changes were seen in paCO₂  or SaO₂. 

The medication regimen used here was helpful in reducing symptoms of dyspnea in these patients.

• The study had a small sample size of less than 30.
• The study had no appropriate control or comparison. 
• Opioids have been shown to reduce dyspnea so as this study was designed, whether the lorazepam in addition was helpful is unclear. 
• Authors mention that patients were anxious, but no measure of anxiety was used or reported.
• All patients were opioid naïve on study entry, so symptoms improving with just the opioids is not surprising.
   

This study adds little new in terms of symptom management for dyspnea. The study design did not help to further define the role of anxiolytics versus opioids for dyspnea management.

The objective of this study was to assess the safety and effectiveness of hydromorphone for the improvement of ventilation and intensity of dyspnea in palliative care patients.

Baseline intensity of dyspnea was recorded at rest and during exertion during a light physical activity. Baseline data, including arterial pressure of carbon dioxide (tcPaCO₂), peripheral oxygen saturation (SaO₂), and pulse frequency (PF) were measured continuously via a noninvasive calibrated digital sensor (i.e., the SenTec Digital Monitor) attached to the patients’ earlobe. They then were initiated on orally administered hydromorphone every four hours and titrated to at least 50% dyspnea reduction. Rescue doses of one-sixth of the calculated daily dose were made available for relief of breakthrough dyspnea.

• The sample was comprised of 14 patients.
• The patients ranged in age from 40–84 years.
• The median age was 64 years.
• Six of the patients were female, and eight of the patients were male.
• The diagnosis documented was advanced terminal cancer.
• Admission diagnoses included lung (7), breast (3), pancreatic (1), colorectal (2), and ovarian (1) carcinoma.
• At admission, all participants reported moderate to severe dyspnea (as indicated by self-reported scores of greater than 3 out of 10 on a numeric rating scale), recorded at rest and on exertion during performance of light physical activity.
• Seven of the 14 patients were not opiate naïve.

The single-site study was conducted in an inpatient setting on a palliative care unit at the Center for Palliative Medicine in Germany.

Patients were undergoing end-of-life and palliative care.

The study was a prospective, nonrandomized trial.

• Respiratory rate
• Subjective rating on intensity of dyspnea numeric rating scale where 0 indicates no dyspnea, 1–3 indicates mild dyspnea, 4–7 indicates moderate dyspnea, and greater than 7 indicates severe dyspnea
• Peripheral oxygen saturation (SaO₂)
• Transcutaneous arterial pressure of carbon dioxide (tcaCO₂)
• Pulse frequency

• Oral hydromorphone was found to significantly decrease dyspnea 120 minutes after use (p < 0.001).
• No increase in tcaCO₂ or decrease in SaO₂ were observed that would suggest respiratory depression.
• Mean respiratory rate decreased from 38.8 breaths (SD = 4.9 breaths) per minute to 34.6 breaths (SD = 4.2 breaths) per minute after 30 minutes and 29 breaths (SD = 3.1 breaths) per minute by 120 minutes.

Use of oral hydromorphone potentially could reduce dyspnea with minimal risk of respiratory depression to patients with advanced or terminal cancer.

• The study had a small sample size, with less than 30 participants.
• This was done for only 120 minutes at a single point in time–longer term efficacy is not clear.
• The level of significance for time to symptom relief was not reported, but effects on severity of dyspnea were apparent as early as 30 minutes after beginning treatment.

Use of hydromorphone in the palliative care setting may serve as an effective treatment alternative for patients with renal impairment or intolerance to morphine in the management of dyspnea and work of ventilation. Hydromorphone may reduce dyspnea even in patients who already are receiving opiates for other symptoms.

The objective of this study was

• To assess whether, and/or in which way, symptomatic treatment of dyspnea with opioids is associated with respiratory depression
• To verify whether nasal water insufflations and opioid administration respectively lead to decrease in dyspnea.

One opioid dose of morphine (mean dose 9.4 mg [SD = 8.8 mg]) or hydromorphone (morphine equivalent dose of 10.8 mg [SD = 3.8 mg])

The sample was comprised of 11 patients with dyspnea (5 severe, 4 moderate, and 2 mild), 8 patients with lung cancer, 2 patients with breast cancer, and 1 patient with acute lymphoblastic leukemia. None of the patients had a history of chronic obstructive pulmonary disease.Two patients were pretreated with opioids for pain control.

The study was conducted on an inpatient palliative care unit.

The study was a prospective, nonrandomized, uncontrolled trial.

• Numerical rating scale (NRS) 0-10 severity of dyspnea
• NRS 0-10 anxiety
• Arterial pressure of carbon dioxide (pCO₂)
• Oxygen saturation %
• Pulse rate

After opioid administration

• No significant change in oxygen saturation (pCO₂) or pulse rate was observed.
• A significant (p = 0.003) decrease in respirations from 41 (SD = 4.7) to 26.5 (SD = 5.3) was noted.
• A significant decrease in dyspnea score (p = 0.003) at rest from 5.4 (SD = 2.5) on admission to study to 0.9 (SD = 0.8) at 120 minutes and dyspnea on exertion from 7.5 (SD = 2.7) to 2.9 (SD = 1.5) at 120 minutes was noted.
• A significant decrease in anxiety also was observed (p = 0.003).

Patients’ ratings showed no significant decrease in dyspnea intensity with nasal oxygen.

Decreased respiratory rate and decreased dyspnea scores with opioid dose were evident.

• Respiratory depression with therapeutic doses of strong opioids was excluded.
• Study limitations are small sample size and uncontrolled, nonrandomized design.
• A possible contamination is that patients also were offered psychological, spiritual, and nonpharmacologic therapy (e.g., breathing therapy, relaxation exercise) as part of routine palliative care.

The purpose of this study was to compare the effects of denosumab to those of zoledronic acid on the pain of patients with advanced breast cancer and bone metastases.

Patients were randomly assigned to receive one of two treatments. One group received monthly subcutaneous denosumab 120 mg and intravenous (IV) placebo. The other group received monthly IV intravenous zoledronic acid with standard dosing and adjustments. Investigators assessed pain severity and the extent to which pain interfered with daily function at baseline and monthly.

• The sample included 2,046 patients.
• Mean patient age was 56.5 years.
• All patients were female.
• All patients had breast cancer with bone metastases. Of all patients,15% had no pain at baseline. Average worst pain severity at baseline was less than five in both groups. At baseline, 16% to 17% of all patients were being treated with strong opioids.

• Multi-site
• Outpatient
• Multi-national

• Late effects and survivorship
• Palliative care 

This study was a double-blind, double-dummy, active, randomized, controlled trial.

• Brief Pain Inventory
• Scale measuring pain severity (The authors defined a meaningful change in pain severity as a change equal to or greater than 2.)

The pain of all patients worsened throughout the study. Compared to patients receiving zoledronic acid who had no or mild pain at baseline, fewer denosumab-receiving patients with the same pain profile progressed to severe pain (p = 0.002). Median time to improvement, among patients reporting worst pain, was similar between groups. Median time to decrease in pain was similar between groups, but patients receiving denosumab waited slightly longer for relief.

Fewer patients in the denosumab group progressed to severe pain than did patients in the zoledronic acid group. Patients in the denosumab group took longer to progress to higher levels of pain.

Both denosumab and zoledronic acid have been shown to reduce skeletal events and bone pain in patients with bone metastases. Findings from this study suggest that denosumab may extend the time that patients have before pain progresses to severe levels. Because denosumab does not require IV administration, it may be a practical alternative for some patients.

To assess the effectiveness of interventions for treatment of oral mucositis or its associated pain for patients receiving chemotherapy or radiation therapy

Databases searched were MEDLINE, CancerLIT, EMBASE, CINAHL, LILACS (Latin American and Caribbean Health Sciences Literature), Cochrane Oral Health Group and PaPaS Trials Registers, Cochrane Central Register of Controlled Trials (CENTRAL), OpenSIGLE, and Current Controlled Trials. Handsearching carried out by the Cochrane Collaboration was included. Reference lists from relevant articles were searched and the authors of eligible trials were contacted to identify trials and obtain additional information.

Search keywords were (neoplasm* OR leukemia OR leukaemia OR lymphoma* OR plasmacytoma OR “histiocytosis malignant” OR reticuloendotherliosis OR “sarcoma mast cell” OR “LettererSiwe disease” OR “immunoproliferative small intestine disease” OR “Hodkin disease”  OR “bone marrow transplant*” OR cancer* OR tumor* OR malignan* OR netropeni* OR carino* or Adenocarcinoma* OR radioth* OR radiat* OR radiochemo* OR irradiat* OR chemo*) AND (stomatitis OR “Stevnes Johnson syndrome” OR “candidiasis oral” OR mucositis OR (oral AND (cand* OR mucos* OR fung*)) OR mycosis OR mycotic OR thrush. Extensive appendices are provided with specific search strategies used for each database. 

Studies were included in the review if they  

• Were randomized controlled trials using placebo, no treatment, or another active intervention.
• Involved patients with cancer receiving chemotherapy or radiotherapy and experiencing oral mucositis.
• Involved any intervention for the treatment of oral mucositis or its associated pain.
• Written in any languages. Papers not in English were translated by members of the Cochrane collaboration.

The final assessment incorporated 32 studies. Out of an initial 95 eligible studies, 64 were excluded because of study design issues, protocol violations, lack of useable data, or no relevant outcomes.

• The final set of studies involved a total of 1,505 patients; 1,023 patients were involved in trials investigating the effectiveness of agents to treat mucositis, and 718 patients were involved in trials evaluating pain relief. 
• Sample sizes ranged from 6–71 patients per treatment or control group.
• Twenty-eight trials included only adult patients, and four included only children.
• Trials included patients treated for a combination of leukemia and solid tumors (n = 14), patients with head and neck cancer (n = 8), and patients who had received bone marrow or stem call transplant (n = 11).

Treatment of mucositis

Summary of data from single trials showed the following interventions to demonstrate statistically significant benefit (p < 0.05).

• Allopurinal mouthwash resulted in improvement in mucositis, eradication of mucositis in some cases, and reduction in time to healing.
• Granulocyte macrophage-colony stimulating factor (GM-CSF) demonstrated mixed results, with two trials showing improved time to healing versus use of providone iodine and antimycotic mouthwash and one trial showing improvement in mucositis by the end of radiotherapy.
• Human placental extract demonstrated improvement in mucositis in one trial.
• Phenytoin mouthrinse was associated with better quality of life than placebo in one trial, but no benefit for pain was found and healing was not evaluated.
• Polyvariant intramuscular immunoglobulin was associated with improvement in mucositis versus placebo in one trial.
• Topical vitamin E was associated with improvement in mucositis and eradication of mucositis compared to systemic vitamin E in one trial.
• Debridement was associated with fewer days to clinical resolution and decreased severity of mucositis, when compared to no debridement.
• Laser treatment was beneficial in management of mild to moderate mucositis compared to sham treatment.

Other interventions for treatment of mucositis evaluated included chlorhexadine versus salt and soda, Gelclair verus sucralfate and mucaine,”Magic” mouthwash versus salt and soda, sucralfate versus placebo and versus salt and soda, and tetrachlorodecaoxide.

Management of pain with mucositis

The following interventions demonstrated statistically significant benefit in managing pain (p < 0.05).

• Opiod use was associated with lower average pain scores when compared to antidepressant use.
• Morphine pharmacokinetically patient controlled analgesia (PKPCA) was associated with lower average pain score than morphine standard patient controlled analgesia (PCA).
• When morphine PCA was compared to continuous morphine infusion, meta-analysis showed no difference in mean pain scores; however, mean opiate intake was reduced with PCA, and PCA was associated with fewer days of pain.

Other findings

• Interventions reviewed that showed no statistical benefit for treatment of mucositis included chlorhexadine, Gelclair, “Magic” mouthwash, and sucralfate.
• Interventions reviewed for management of associated pain that demonstrated no statistical benefit included hydromorphone PCA versus morphine, Alfentanil versus morphine, Dicofenic versus placebo, PCA versus staff controlled, hypnosis, relaxation, and imagery.
• Out of 27 different interventions evaluated for treatment of mucositis, only one comparison was significant for one outcome: low level laser treatment reduced the severity of mucositis.
• No evidence was found to suggest a difference in pain control between continuous infusion and PCA; however, the PCA group required less morphine, and the pain lasted two less days.

• Some evidence exists that low level laser treatment may help reduce severity of mucositis.
• No evidence suggests that PCA is more effective than continuous infusion for controlling pain. Weak evidence is available to support that PCA is associated with less opiate used per hour and that the duration of pain may be reduced.
• No clear benefit appears to be associated with antimicrobial use and GM-CSF for prevention or management of mucositis.
• This review demonstrated weak and unreliable evidence of benefit for interventions for mucositis.

The lack of independent duplication of studies investigating the same intervention limits the strength of evidence and ability to generalize results.

Most studies reviewed had small sample sizes and may have been underpowered to demonstrate significant differences in outcomes.

Different scoring systems for mucositis were used, and, in some studies, the method of scoring was not defined.

The need for further well-designed trials to evaluate the effectiveness of interventions continues.

Adoption of standard clinical outcome measures should be considered, including patient-based measures and inclusion of the cost of interventions.

The music-listening intervention included preferred music and recommended relaxation techniques (e.g., progressive muscle relaxation, imagery, and positive self-talk). Sessions were guided by a music therapist on a 90-minute cassette. Frequency of listening (dose) varied, but duration was not reported.

• The sample was comprised of 63 patients with cancer at varied stages undergoing curative radiotherapy (mean age = 57.8 years).
• Of the patients, 61.9% were male and 85.7% were non-Hispanic White.
• Patients were included if they were aged 18 years or older, were receiving at least three radiotherapy treatments, and were able to read English.
• Patients were excluded if they had a documented psychiatric illness, cognitive impairment, or current treatment for a brain tumor.

The study was conducted at a comprehensive cancer center, including a Veteran's Administration facility.

Not specified

This was a randomized trial in which patients were assigned to one of the two groups:  the music therapy group (n = 35) or the control group (n = 28).

• Profile of Mood States (POMS), fatigue and vigor subscales combined
• Hospital Anxiety and Depression Scale (HADS)
• Distress Numeric rating scale
• Pain Numeric Rating Scale
• Measures were taken weekly.

Fatigue increased from baseline to the end of treatment for both groups. No significant difference in intervention effects was observed. Fatigue, depression, pain, or anxiety showed a significant relation between frequency of listening and emotional distress. Higher use of music was associated with greater reduction in treatment-related stress.

• The study had a small sample size.
• Variable listening dose and relaxation are both confounding factors with music therapy.

Referral to a music therapist is necessary for the delivery of the intervention.

To determine the effectiveness of biosimilar granulocyte colony-stimulating factor (G-CSF) as compared to originator G-CSF in reducing the duration of neutropenia following autologous peripheral blood stem cell transplantation (APBSCT).

The study group received biosimilar G-CSF following myeloablative chemotherapy and APBSCT. G-CSF was initiated when the absolute neutrophil count (ANC) dropped below 0.5 and continued until the ANC exceeded 1.5 for three consecutive days. Hematopoietic recovery was compared to the control group, which had received originator G-CSF.

• N: 46 (23 in prospective cohort)
• AGE: Study group: 47 (SD = 13); control group: 53.6 (SD = 13)
• MALES: 52%   
• FEMALES: 48%
• KEY DISEASE CHARACTERISTICS: Hematological malignancy (multiple myeloma, Hodgkin’s lymphoma, non-Hodgkin’s lymphoma, and acute myelogenous leukemia)

• SITE: Single site    
• SETTING TYPE: Inpatient    
• LOCATION: Lublin, Poland

• PHASE OF CARE: Active antitumor treatment

Prospective observational with historical control comparison

The primary comparator studied was duration of G-CSF treatment. Adverse events were also compared.

There was no significant difference between the biosimilar and originator G-CSF groups with respect to duration of therapy (p=0.43). The frequency of occurrence of the most common adverse events (neutropenic fever and bone pain) were also comparable.

Biosimilar G-CSF had been previously demonstrated to have similar efficacy and safety as originator G-CSF. This study confirms the value of using biosimilar G-CSF in a post-transplantation setting.

• Small sample (< 100)
• Risk of bias (no blinding)
• Risk of bias (no random assignment)

Biosimilar G-CSF therapy has similar efficacy to originator G-CSF in patients with ABPSCT, but with significant cost savings. Because patients and caregivers are typically unable to work during the prolonged transplantation process, financial stressors are a frequent concern. Lowering medical costs helps alleviate financial concerns for patients and insurers.

To evaluate the effects of the early onset rehabilitation program on shoulder mobility, functional capacity, lymphedema, and postoperative complications in patients who had modified radical mastectomy

Patients were randomly assigned to either the treatment group (n = 27) or home exercise program group (n = 30). In the treatment group, specific shoulder-hand-elbow range-of-motion exercises were performed under the supervision of a physiotherapist until the drains were out, then participants followed a physiotherapy program for eight weeks. In the home exercise program group, patients received a form that demonstrated how to perform the exercises by themselves after removal of the drains. Each exercise was taught by a physiotherapist until the exercise was performed properly. Both groups were informed about skin care and other issues that they should pay attention to during daily living activities. Each patient was assessed preoperatively and then postoperatively on the fifth day and one, three, and six months after by another physiotherapist who was blinded to the groups of the patients.

• The sample size consisted of 57 female patients with 27 in the intervention group and 30 in the comparison group.
• Patients were about 50 years of age.

The setting was a single site in Turkey that included inpatient and outpatient.

The study used a randomized controlled trial design.

• Circumferential measurements were used to assess lymphedema.
• Range of motion of the operation-side shoulder joint was measured by Myrin goniometry while the patients were sitting in a chair.
• A functional questionnaire was used.

There were statistically significant time-related changes in all range-of-motion measurements and functional questionnaire scores in both groups. The differences over time in flexion, abduction, and adduction movements were significantly better in the treatment group compared with the home exercise program group  (p < 0.01, p < 0.001, p < 0.005, respectively). The mean range of flexion and abduction returned to almost preoperative values more quickly in the treatment group compared with the home exercise program group. The recovery of upper-extremity functional questionnaire score was also significantly better in the treatment group compared with the home exercise program group  (p < 0.05). There was no statistically significant variance in circumferential difference between the groups.

Early rehabilitation started on the first postoperative day did not have an adverse effect on local infection, hematoma, and seroma formation and did not cause an increase in duration and amount of lymphatic drainage.

• The sample was small (N < 100).
• The follow-up time for the late effect of breast cancer treatment, lymphedema, was shorter (only six months).

Nurses and clinicians should remember to refer patients with breast cancer to a rehabilitation specialist. During the postoperative period, patients should be closely monitored to increase their adaptation and compliance to an early onset exercise program.